Festa v The Queen

I do not intend in what follows either to denigrate the current script or to boo the players. I wish, rather and simply, and speaking as one who is called upon to preside regularly over criminal trials with a jury, to draw attention to some practical results for trial Judges of the matters which it has been necessary to discuss in the context of the present particular appeal.

In this State a Judge who is presiding at a criminal trial comes to the summing-up burdened with a great deal of appellate judicial advice as to how the summing-up should be framed. Three particular examples will make the point.

It seems to me, if I may say so with unfeigned respect, that the combined effect of the decisions in Longman , in Crampton and in Doggett makes it, if not quite impossible, at least extremely and unnecessarily difficult for a conscientious trial Judge when directing a jury, to give dutiful effect, as of course he must do, to the requirements of those decisions, while simultaneously giving effect to the requirement that he be succinct, simple and clear.

Suppose that the case is one like Doggett itself, that is to say, a case in which there is some, albeit perhaps patchy, corroboration of at least some parts of the complainant's own evidence. Suppose that counsel for the accused is alive to the forensic problems that are described by Gleeson CJ in paragraph [9] of his Honour's judgment in Doggett . Suppose that counsel for the accused, mindful of that forensic minefield, does not ask for a Longman direction; or, indeed, asks that no such direction be given. Where, it might reasonably be asked, does that leave the trial Judge? There are, obviously I should have thought, only two practical choices. One is to accede to the request that there be no Longman direction. The other is to over-ride that request in the spirit of Pemble v The Queen (1971) 124 CLR 107. If the Judge takes the former course, the current state of the law makes it overwhelmingly probable that any resulting conviction will be upset on appeal. If the Judge takes the Pemble approach, then he will find himself mired inevitably in the very problems of which Gleeson CJ gives warning in the paragraph [9] to which I have earlier referred; and there is raised thereby the prospect that a requirement intended to protect the accused turns out to have in fact exactly the opposite practical result.

Suppose, to take another practical possibility, that the trial Judge gives a Longman direction. Suppose that the jury, after retirement to consider its verdict, sends in a note asking the trial Judge whether there is any, and if so what, evidence indicative of any practical difficulty or difficulties in fact encountered by the particular accused by reason of the effluxion of time between the time of the alleged offence and the time of complaint. And suppose that, as in the present particular case, there is in truth not a scintilla of evidence of any such practical forensic disadvantage having been suffered by the particular accused. What, then, is the trial Judge to do? If he tells the jury the simple truth, it is likely to be held that he has thereby destroyed the efficacy of the Longman warning. If he tells the jury that it is no concern of theirs to enquire whether or not there is any such particular evidence; but that they are to proceed upon the simple basis that the experience of the Courts suggests that there has been some such forensic disadvantage, whether or not the particular accused has claimed it, or given evidence of it, the practical likelihood is that the jury will reason that, since it is not possible to see in any concrete way evidence of actual forensic disadvantage, there is no good reason why, as a matter of fact, it should be found that there has been any. In which event the apparent lack in the particular case of any discernible factual underpinning for the Longman direction produces, once again, the result that an intended protection for the accused ceases to afford the accused any protection, and tends rather to have the opposite effect in practice.

And, quite apart from the matters raised in the two immediately preceding paragraphs, what guidance, it might fairly be asked by a trial Judge, does the current state of the law give as to the practical definition of the sort of time lapse which justifies, in a real sense, the giving at all of a Longman direction. It is, I should think, simple enough for any trial Judge to grasp the concept, articulated by Kirby J in Doggett , that: "(t)he criterion for the provision of a warning as stated in Longman is not mathematically precise" . But what exactly, the trial Judge might find himself wondering, is meant by the proposition that "in a case involving a comparatively short interval between the alleged offence and a subsequent complaint to family members or to authorities, a warning might not be necessary" (emphasis added)? Even after some definition has been given to the fluid concept of "a comparatively short interval" , it remains necessary to consider, granted that there has been such a short interval, not whether a Longman direction might or might not be required, but whether such a direction is or is not required. If the Judge approaches that conundrum upon the basis that the determining consideration is the presence of an unacceptable risk of unfairly prejudicial forensic disadvantage, then either the Judge assumes as a matter of law, and notwithstanding the complete absence of relevant evidence, that there has been some such forensic disadvantage; or the Judge refuses to give the direction unless there is in hand at least some credible evidence from which there may be drawn a reasonable inference that there has been, in a sense that is not far-fetched or fanciful, relevant forensic disadvantage. Either way, the answer gives rise, as I have attempted to demonstrate in the preceding paragraphs, to yet further practical problems.

There is one additional matter worthy of present consideration. A commonsense understanding of the real world suggests that a jury which is given a Longman direction in the form now apparently required, is likely to reason that the trial judge, although he has stressed repeatedly that it is not for him to tell the jury how the facts should be found, is in fact sending a none-too-subtly coded indication to the jury that the dangers of convicting are such that the jury ought to return a verdict of not guilty. 116 Wood, CJ. at CL. added some observations of his own and said, dealing with passages in Longman (supra), Doggett (supra) and Crampton (supra):- These passages have been taken up, so it seems to me, as requiring that an instruction in equally positive terms, be given in every case involving a substantial delay, irrespective of whether or not there is any evidence, or basis beyond suspicion, that the absence of contemporaneity between the alleged offence and complaint, or trial has in fact (not 'might have') denied to the accused a proper opportunity to meet the charge or charges brought: see for example R v Roddom (2001) NSWCCA 168, R v GJH (2001) NSWCCA 128 and R v Roberts (2001) NSWCCA 163.

Put another way, the effect of these decisions has been to give rise to an irrebuttable presumption that the delay has prevented the accused from adequately testing and meeting the complainant's evidence; and that, as a consequence, the jury must be given a warning to that effect irrespective of whether or not the accused was in fact prejudiced in this way. 117 His Honour identified the reasoning in those decisions as deriving the direction from the onus and standard of proof when he reflected on the form of the direction usually given and called for a re-consideration of the form of the warning:- The difficulty which I have with this proposition is that it elevates the presumption of innocence, which must be preserved at all costs, to an assumption that the accused was in fact innocent, and that he or she might have called relevant evidence, or cross examined the complainant in a way that would have rebutted the prosecution case, had there been a contemporaneity between the alleged offence and the complaint or charge. That consideration loses all of its force if, in fact, the accused did commit the offence. In that event there would have been no evidence available of a positive kind, relating for example to the existence or ownership of the premises, or of a motor vehicle or other item, associated with the offence charged, or going to establish an alibi for the relevant occasion, no matter how contemporaneous the complaint or charge was with the offence.

I do not seek by these observations to question the common sense considerations that delay can, and in many cases will, occasion difficulty to an accused who may have lost an opportunity to investigate whether a complaint of sexual misconduct could, or could not, have occurred at those places, or within the time frames alleged and, even more importantly, to gather physical evidence in the form of records, or evidence from other witnesses that would support such a proposition. The simple fact is that records can be lost or mislaid, and the memory of potential witnesses may dim, or they may not be capable of being found.

For these reasons, as Spigelman CJ noted in R v Johnston (1998) 45 NSWLR 362, in sexual assault cases which commonly turn upon a direct conflict or word against word, between a complainant and an accused, and where there is no independent verification of whether the event occurred, or of whether it was consensual, the contextual details of the alleged conduct will often prove to be " of definitive significance " (at p 370).

These are matters which can and should properly be placed before a jury, as relevant to its assessment of the prosecution case. My concern lies more with the unequivocal nature of a warning , which must be given by a trial judge who does not himself or herself know where the truth lies, that the accused was unable to adequately test and meet the prosecution case.

If in fact the accused did commit the offence charged, any such warning or direction would be misleading if not positively untrue. A similar consideration would apply if, in fact, the accused was not prejudiced, being able, in the particular circumstances of the case, for example to call positive and objective evidence in rebuttal, or where the absence of contemporaneity did not in any way deprive him or her of such an opportunity.

The latter circumstance might arise, for example, where the complaint related to a time and place where the accused was in fact living alone with the complainant, and in circumstances where, no matter what inquiries were made, the case became one of word against word, such that rebuttal evidence could never have been obtained.

I have no difficulty with the decision in Longman , or with the subsequent decisions, if understood as decisions on their own facts, that is, as cases where the delay was of such an inordinate degree that it was impossible for an accused even to begin to investigate the circumstances alleged. Nor do I have any problems with the warning now required, where there is evidence, or good reason to suppose positively, that the accused has been prejudiced.

Otherwise, I would respectfully raise for consideration, whether some modification should be made to the warning which, it seems to me, must now be given, in invariable and emphatic terms, of the fact of prejudice. In that regard I do not consider it unimportant that, in the various passages cited, the reasons given for the warning have at times spoken in terms of the possibility, rather than the fact, of prejudice, as underpinning it (see also in this regard the observations of Gaudron, McHugh and Gummow JJ in Jones v The Queen (1997) 191 CLR 439 at 454 to 455.

Nor do I consider it irrelevant that the impact of delay is double edged, since it is just as likely to occasion practical difficulty for the prosecution in pinning down times and places, and in gathering relevant witnesses.

For the reasons identified in Regina v GPP ([2001] NSWCCA 493), and by Sully J in this appeal, notwithstanding the number of times that the Longman direction has been considered, there still is a sharp division in the High Court (following Doggett v The Queen ) as to the circumstances when, and the terms in which, the necessary warning should be given. Additionally, there is the concern upon which I have focused concerning whether or not the warning should be given in terms that the appellant has been forensically disadvantaged on the one hand, or might have been so disadvantaged, on the other hand.

It needs to be borne in mind that the direction presently under consideration is but one of a multitude of directions which now fall to be considered by a trial judge faced with the task of summing up to a jury in a sexual assault case. They include:- (a) the Murray direction ( R v Murray (1987) 11 NSWLR 12) to the effect that where there is only one witness asserting the commission of a crime, the evidence of that witness " must be scrutinized with great care " before a conclusion is arrived at that a verdict of guilty should be brought in; (b) The Longman direction (as reinforced in Crampton and Doggett ), that by reason of delay, it would be " unsafe or dangerous " to convict on the uncorroborated evidence of the complainant alone, unless the jury scrutinizing the evidence with great care, considering the circumstances relevant to its evaluation and paying heed to the warning, were satisfied of its truth and accuracy. (c) The Crofts direction ( Crofts v The Queen (1996) 186 CLR 427), if a jury is to be informed, in accordance with s 107 of the Criminal Procedure Act , that a delay in complaint does not necessarily indicate that the allegation is false, and that there may be good reasons why a victim of sexual assault may hesitate in complaining about it, then it should also be informed that the absence of a complaint or a delay in the making of it may be taken into account in evaluating the evidence of the complainant, and in determining whether to believe him or her (but not in terms reviving the stereotyped view that complainants in sexual assault cases are unreliable or that delay is invariably a sign of the falsity of the complaint: Crofts at 451). (d) The KRM direction ( KRM v The Queen (2001) 75 ALJR 550) to the effect that, except where the evidence relating to one count charging sexual assault is admissible, in relation to another count or counts alleging a separate occasion of such an assault, the jury must consider each count separately, and only by reference to the evidence which applies to it; balancing that direction, where appropriate, by a reminder that if the jury has a reasonable doubt concerning the credibility of the complainant's evidence on one or more counts, they can take that into account when assessing his or her reliability on the other counts (see Regina v Markuleski [2001] NSWCCA 290 at paras 259-263). (e) Any warning which may be required by reason of a ruling that limits the use of evidence concerning a complaint, or delay in complaint, to the question of credibility (eg under s 108(3) of the Evidence Act as an exception to the credibility rule), or alternatively that allows it to be taken into account (under s 66 of the Evidence Act as an exception to the hearsay rule) as evidence of the facts asserted. (f) The Gipp warning (conveniently so called, although there was divided reasoning in Gipp v the Queen (1998) 194 CLR 106) concerning the way in which evidence of uncharged sexual conduct between an accused and a complainant can be taken into account as showing the nature of the relationship between them, but not so as to substitute satisfaction of the occurrence of such conduct for proof of the act charged; (g) Any warning that may be necessary in relation to the use of coincidence evidence (under s 98 Evidence Act ) where the accused is charged in the one indictment with sexual assault against two or more complainants, requiring the jury to be satisfied beyond reasonable doubt first of the offences alleged in respect of one complainant, and then of the existence of such a substantial and relevant similarity between the two sets of acts as to exclude any acceptable explanation other than that the accused committed the offences against both complainants; (h) A BRS direction ( BRS v The Queen (1997) 191 CLR 275) that where evidence revealing criminal or reprehensible propensity is admitted, but its use is limited to non propensity or tendency purposes, for example those considered proper in that case, then it is to be used only for those purposes and not as proof of the accused's guilt.

In all of these circumstances, bearing in mind also the desirability of containing a summing up to an acceptable length; of ensuring its immediate relevance to the actual trial ( Regina v Zorad (1991) 19 NSWLR at 91, Regina v Williams NSWCCA 10 October 1990, KRM v The Queen, and R v Chai [2002] HCA 12 at 18 ); of avoiding unnecessary judicial input into the fact finding process ( Azzopardi v The Queen [2001] HCA 24 ); and of paying due respect to the tactical considerations which are best judged by trial counsel in the actual atmosphere of the trial, I consider it timely for there to be a further review of the evidentiary, and other requirements of procedural law that apply to cases of sexual assault, particularly those involving children.

In summary, in relation to this specific aspect of this appeal, I repeat my concern that the direction currently required amounts to a warning that, in fact, the accused was unable by reason of the delay to test or to meet the prosecution case. Two problems emerge:- (a) the giving of such a direction involves a finding of fact that this was the case, whereas it might not be so - indeed it may be positively incorrect; and (b) so far as it involves a finding of fact then it intrudes into the proper role of the jury. 118 Although these passages cited from both cases are extensive, their importance is such as to warrant reiteration. With the observations of all of their Honours I have cited, I agree. I further observe that where the warning, as here, neglects to draw to the jury's attention that the delay has or may have prevented the testing of the Crown case, there is another side to the commonsense observation of Sully, J. which also involves a judicial intrusion into the jury's function, where such direction as has been given focuses upon the effect of delay on the accused's case rather than the Crown case. In such a case the Crown case may be too readily accepted by the jury, and because the accused is or may be unable to meet it (as the jury have been judicially informed is the case) the conclusion of guilt might be more easily drawn. The absence of an appropriate warning may cause a jury to apply a lesser standard of satisfaction. In my view, an appropriate warning will refer to the possible effects on both the Crown and defence cases and will refer to the difficulty of testing the Crown case and the assessment of the weight to be given to it. 119 The warning is intended to ensure trials are fair which might otherwise have been stayed as abuses for delay. Analysed in this way, an appropriate warning is a legal necessity for a fair trial. The absence of an adequate direction will result in the trial not being a fair trial according to law (see Mraz v. Regina (1955) 93 CLR 493). In such circumstances, it is hard to see room for the application of Rule 4 or the proviso. 120 Heydon, JA. in GPP (supra) regarded the absence of an appropriate direction as not "fundamental" in the sense referred to in Regina v. Wilde (1988) 164 CLR 365 and Regina v. Glennon (1994) 179 CLR 1 but as reflecting on whether a conviction might be regarded as inevitable that is, whether the direction would found a basis for the possible experiencing of a reasonable doubt. Such a view relates the necessity for a Longman direction to the onus and standard of proof. Whether its absence would cause such a fundamental defect in the trial as is referred to in those cases need not be resolved since nonetheless its absence would result in a trial which was not a fair trial according to law. Thus unless, postulating a proper direction, the appellate court would have no reasonable doubt of guilt on the evidence at trial, the appeal should be upheld (see Festa v. The Queen (2001) 76 ALJR 291). 121 In my view, even though no application for a proper direction is made (unless perhaps, the lack of application was established as proceeding from a tactical consideration), even where, as here, conviction on all or most of the counts in both trials would be inevitable or nearly inevitable, the present state of the law requires that unless there is a direction which approaches nearly enough to the Longman requirements so that the jury could be seen to have the requisite understanding the appeal should be allowed. 122 It is for these reasons I have concluded the appeals should succeed. The evidence is, however, such that there should be new trials. I therefore agree with the orders proposed by Kirby, J. 123 KIRBY J: I have had the advantage of reading the judgment of Hodgson JA in draft. There were two broad areas of complaint in respect of each trial. They were: · First, the admission of evidence, either upon the basis of relationship or coincidence, and the directions given in respect of such evidence. · Secondly, the directions given in respect of delay; the so-called Longman Direction. 124 I agree with Hodgson JA, and for the reasons he states, that there was no error in respect of the first issue in either trial. I do not, however, share his Honour's view on the second issue. I shall, therefore, deal with that issue.

The Longman Direction 125 Both trials involved significant delay. It is necessary to look at what was said on the question of delay in each trial. 126 The first trial concerned the allegations of CPS. The appellant (WRC) was indicted on 28 August 2000 upon four counts. The offences were alleged to have been committed at various times between 1 October 1976 and 30 June 1977. CPS was born on 27 August 1968. The appellant was his teacher in Year 3. The proceedings, therefore, were based upon the recollection of CPS, now an adult, of incidents which occurred when he was a child aged approximately eight years. The complaint was first made to the police in 1997, more than 20 years after the events described. The trial, in August 2000, was more than 23 years after the alleged offences. 127 His Honour's directions on this issue, in the first trial, were brief. He said this: There is also one other matter that I bring to your attention, in which the law requires me to bring to your attention and it is a matter that has been also referred to you by counsel for the accused. And this is: that because of the delay between the time when these events are alleged to have occurred - that is in 1977 in relation to [CPS] and 1978 in relation to [TEL] - and the time of the allegations were brought to police notice, which occurred in 1998, following the police being brought into the matter with the police statement - do you recall the evidence of the police statements being made in January 1997? - following that delay of which is twenty-two or twenty-three years, or thereabouts, the accused has necessarily had difficulties in meeting those allegations.

The examples of the difficulties that were brought to your attention during the course of the trial and the closing addresses are, in a sense, the accused has had to conduct his defence in relation to the allegations that he indecently assaulted [CPS] and [TEL] at the army barracks in 1977 and 1978. And now, after a period of years, [there] have been changes made in some respects to the barracks. An example, of course, is after twenty-two years there is a disputed issue as to what type of floor was on the drill hall; a dispute about what kind of windows were in the drill hall (painted or frosted); what was the situation in relation to an occasion, was it a parade day when the squadron was being presented to 100 to 150 people milling around; or was it an occasion when the squadrons were not there and there was only a skeleton staff, perhaps a dozen, or in some cases perhaps two or three.

I am just drawing your attention to these issues which have arisen in this trial, and the fact that you, the jury, are asked to reflect on and consider findings in relation to disputed issues that deal with matters that are some twenty-two or twenty-three years in the past. 128 The second trial was the consequence of allegations by TEL. The appellant was indicted on 30 October 2000 upon seven counts. The offences were said to have taken place at various times between 18 February 1978 and 28 January 1979. The complainant was born on 10 March 1969. He went to the same school as CPS. The appellant was his teacher. The complainant first made a complaint in 1979 to the school authorities. He made a statement to the police in 1997. TEL was, at the time of the alleged offences, about nine years old. The trial took place more than 21 years after the events described. 129 In the second trial, his Honour summed up on this issue in these terms: There is one further direction that I give you in relation to this matter and that is this, that because of the delay between the time when these events are alleged to have occurred, that is in 1978, and the time when these allegations were brought to ... notice, it (sic) necessarily had difficulty in meeting these allegations.

That is a matter, members of the jury, that I am bound under the law to bring to your attention and to remind you of those difficulties.

It is a matter that you should take into consideration, that is the prejudice that has been suffered by the accused in meeting these claims, meeting these allegations, that his ability to recall and gather evidence pointing to his innocence has been adversely affected.

That is a matter for you to consider. It is a matter for you to take into account, as I say, that he has been placed in a difficult situation, now having to answer allegations relating to 1978.

Now ... the accused has been placed in a situation of hardship in adequately and properly preparing his defence by reason of that delay by reason of the fact that the matters that might have been fresh in the memory in 1978 might well have faded from memory. And there may be other matters that could have been attended to but were not.

And in this case, you heard from Mr Webb, counsel for the accused, refer to the difficulties in relation to a number of disputed issues such as was there a stencilling room in the ... Prep School in 1978 on the first floor.

And the difficulty in the year 2000 of ascertaining what the position was. Did he have a mustardy coloured four door sedan in 1978, or did he not. Was the school playing cricket on Saturday at Primrose Park or not. And there are other matters that you have been told about, and they are matters, members of the jury, that you should reflect on and consider in relation to the fact that there has been an absence of complaints. 130 Each case, therefore, concerned allegations of sexual molestation of a young child. One, at least, involved delay in making any complaint. Both cases involved significant delay in prosecution. A Longman warning was called for in each case. The Crown did not suggest otherwise. 131 The appellant complains that, in neither trial, did the directions conform with the requirements of Longman v The Queen (1989) 168 CLR 79. The deficiencies asserted by the appellant, common to both trials, were as follows: · First, the direction was not given with the firm voice and authority of the Judge administering a warning, identifying matters which the jury was bound to take into account. Rather, the issues were put in a way which suggested they were a summary of points made by counsel. · Secondly, the appellant's difficulties, caused by delay, were not adequately identified. The only matter identified, namely, the physical changes to the army barracks, was an aspect which was relatively minor. · Thirdly, no reference was made to the need for extreme caution before acting upon the recollection of events that occurred in early childhood. 132 A number of issues arise: · First, what is required by a Longman warning? · Secondly, were those requirements met, or substantially met, in the directions given in each trial? · Thirdly, if they were not, is there any room for the application of Rule 4, or the proviso?

The Requirements of Longman 133 The statement of principle in respect of the obligation to warn juries comes, of course, from the joint judgment of Brennan, Dawson and Toohey JJ in Longman v The Queen (supra), where their Honours said this: (at 91) But there is one factor which may not have been apparent to the jury and which therefore required not merely a comment but a warning be given to them: see Reg v Spencer [1987] AC 141. That factor was the applicant's loss of those means of testing the complainant's allegations which would have been open to him had there been no delay in prosecution. Had the allegations been made soon after the alleged event, it would have been possible to explore in detail the alleged circumstances attendant upon its occurrence and perhaps to adduce evidence throwing doubt upon the complainant's story or confirming the applicant's denial. After more than twenty years that opportunity was gone and the applicant's recollection of them could not be adequately tested. The fairness of the trial had necessarily been impaired by the long delay (see Jago v District Court (NSW) (1989) 168 CLR 23) and it was imperative that a warning be given to the jury. The jury should have been told that, as the evidence of the complainant could not be adequately tested after the passage of more than twenty years, it would be dangerous to convict on that evidence alone unless the jury, scrutinizing the evidence with great care, considering the circumstances relevant to its evaluation and paying heed to the warning, were satisfied of its truth and accuracy. To leave a jury without such a full appreciation of the danger was to risk a miscarriage of justice. The jury were told simply to consider the relative credibility of the complainant and the appellant without either a warning or a mention of the factors relevant to the evaluation of the evidence. That was not sufficient." 134 The need for such a warning can be traced to two matters (cf Kirby J, Doggett v The Queen (2001) 182 ALR 1 at 26). They are matters known to the Courts through experience, but unlikely to be known to juries, unless the subject of a warning. The first matter was described by Gaudron, Gummow and Callinan JJ in Crampton v The Queen (2000) 176 ALR 369, in these words: (at 379) ... the denial to an accused of the forensic weapons that reasonable contemporaneity provides, constitutes a significant disadvantage which a judge must recognise and to which an unmistakable and firm voice must be given by appropriate directions. 135 Their Honours said: The trial judge should have instructed the jury that the appellant was, by reason of the very delay, unable adequately to test and meet the evidence of the complainant. 136 In Doggett v The Queen (2001) 182 ALR 1, Gaudron and Callinan JJ elaborated in these words: (at 11) ... the problems with which Longman is intended to deal are not confined to difficulties of recollection that the passage of time might cause for an accused. Of equal, and in some cases of which this might be one, of more importance is the denial by the effluxion of time, to an accused of the forensic weapons that a timely complaint might allow an accused to assemble, such as evidence as to where he was or what he was doing, or what other potential witnesses were doing when the offences were alleged to have occurred." 137 The second matter emerges from the judgments of Deane and McHugh JJ in Longman. McHugh J said this: (at 107) The fallibility of human recollection and the effect of imagination, emotion, prejudice and suggestion on the capacity to 'remember' is well documented. 138 His Honour added: (at 107/108) 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. 139 The first of these matters has been questioned by this Court in R v BWT [2002] NSWCCA 60 (Wood CJ at CL, Sully and Dowd JJ). In that case the trial Judge gave a warning which was framed in terms of the accused possibly having difficulties occasioned by delay ("it may have resulted in some difficulties for the accused in defending himself"), rather than the accused in fact having such difficulties. Applying Longman, the Court of Criminal Appeal felt obliged to uphold the appeal. Had Longman (and the cases after it), been expressed less emphatically, as their Honours clearly believed appropriate, then the appeal would have been dismissed. Wood CJ at CL said this: (para 13/15) 13. These passages have been taken up, so it seems to me, as requiring that an instruction in equally positive terms, be given in every case involving a substantial delay, irrespective of whether or not there is any evidence, or basis beyond suspicion, that the absence of contemporaneity between the alleged offence and the complaint, or trial has in fact (not 'might have') denied to the accused a proper opportunity to meet the charge or charges brought: see for example R v Roddom [2001] NSWCCA 168, R v GJH [2001]NSWCCA 128 and R v Roberts [2001] NSWCCA 163.

14. Put another way, the effect of these decisions has been to give rise to an irrebuttable presumption that the delay has prevented the accused from adequately testing and meeting the complainant's evidence; and that, as a consequence, the jury must be given a warning to that effect irrespective of whether or not the accused was in fact prejudiced in this way.

15. The difficulty which I have with this proposition is that it elevates the presumption of innocence, which must be preserved at all costs, to an assumption that the accused was in fact innocent, and that he or she might have called relevant evidence, or cross examined the complainant in a way that would have rebutted the prosecution case, had there been a contemporaneity between the alleged offence and the complaint or charge. That consideration loses all of its force if, in fact, the accused did commit the offence. In that event there would have been no evidence available of a positive kind, relating for example to the existence or ownership of the premises, or of a motor vehicle or other item, associated with the offence charged, or going to establish an alibi for the relevant occasion, no matter how contemporaneous the complaint or charge was with the offence. (emphasis in original) 140 His Honour recognised that, as a matter of common sense, delay can, and in many cases will, occasion difficulties to an accused. However, Wood CJ at CL identified his difficulty with Longman, as it has been interpreted, in these words: (para 18) These are matters which can and should properly be placed before a jury, as relevant to its assessment of the prosecution case. My concern lies more with the unequivocal nature of a warning , which must be given by a trial judge who does not himself or herself know where the truth lies, that the accused was unable to adequately test and meet the prosecution case" (emphasis in original) 141 However, with respect, I disagree. Significant delay does create disadvantage, guilty or innocent. Our system of criminal justice seeks to fireproof each conviction against error. It does so through the combined operation of the fundamental principles of criminal justice (the presumption of innocence and the requirement of proof beyond reasonable doubt) and the adversarial system. The accused may cross examine and call evidence. The proof offered by the Crown must be sufficiently robust to withstand the assault upon it even of the guilty. 142 The potency of cross examination as a means of exposing unreliable evidence diminishes significantly with time. In State v Saporen (1939) 205 Minn 358 (quoted in McCormick on Evidence, 2nd Ed (1972), page 602), the following was said: The chief merit of cross-examination is not that at some future time it gives the party opponent the right to dissect adverse testimony. Its principal virtue is the immediate application of the testing process. Its strokes fall while the iron is hot. False testimony is apt to harden and become unyielding to the blows of truth in proportion as the witness has opportunity for reconsideration and influence by the suggestions of others. ... 143 The method of cross examination is an attack upon detail, exposing contradictions and unreliability. If the trial is undertaken within a reasonable time, the excuse for unreliability of fading memory is unlikely to be persuasive. However, where the trial is delayed, the accused is disadvantaged in two ways. First, the testimony is likely to be more vague, bereft of the detail which may be used to expose unreliability. Secondly, that absence of detail, and any contradiction that may happen to emerge, can the more easily be explained by reference to the passage of time. The jury therefore is more likely to be forgiving of shortcomings in the complainant's evidence, especially in the context of charges which arouse strong feelings of prejudice or revulsion (cf Kirby J Doggett (supra) at 25, para 118). 144 Returning to the requirements of Longman, Sully J in R v BWT (supra) provided a helpful analysis of matters which must be included to conform with the authorities. His Honour said this: (para 95) 95[3] ... Not only must the direction be given; but it must be cast in a form that manifests, and is seen plainly to manifest, certain characteristics which one can draw, as follows, from the various statements of principle in Longman itself and in the subsequent decisions in Crampton and in Doggett : (a) The direction must be cast in the form of a warning . Any form of expression which is thought to have the character of a comment , or even of a caution will not sufficiently comply with what is required by law.

It seems to me to follow that any trial Judge who is framing a Longman direction ought to ensure that the direction is framed, in terms, as a warning .

(b) That which is to be warned against is, to return to the majority judgment in Longman itself: 'that, as the evidence of the complainant could not be adequately tested after the passage of [the particular period relevant in the particular trial], it would be dangerous to convict on that evidence alone unless the jury, scrutinising the evidence with great care, considering the circumstances relevant to its evaluation and paying heed to the warning, was satisfied of its truth and accuracy' . (emphasis in original) 145 His Honour then distilled six matters which a Longman direction should include. He said this: (para 95) 95[3] ... a trial Judge who is framing a Longman direction must ensure that the final form of the direction to the jury covers in terms the following propositions: first, that because of the passage of time the evidence of the complainant cannot be adequately tested; secondly, that it would be, therefore, dangerous to convict on that evidence alone; thirdly, that the jury is entitled, nevertheless, to act upon that evidence alone if satisfied of its truth and accuracy; fourthly, that the jury cannot be so satisfied without having first scrutinised the evidence with great care; fifthly, that the carrying out of that scrutiny must take into careful account any circumstances which are peculiar to the particular case and which have a logical bearing upon the truth and accuracy of the complainant's evidence; and sixthly, that every stage of the carrying out of that scrutiny of the complainant's evidence must take serious account of the warning as to the dangers of conviction .

The Directions in these Cases 146 The directions given in the first trial, in my view, plainly did not conform, or substantially conform, to these requirements. The direction was not framed as a warning. Arguably, the direction adequately dealt with the difficulties which the accused experienced in presenting his own case. The jury was told that the accused necessarily had difficulties in meeting the allegations by reason of the delay. An illustration was provided concerning the appearance of the army barracks, which had changed. However, the jury was not told that the complainant could not be adequately tested by reason of the delay. Nor was reference made to many of the other issues identified by Sully J. 147 The direction given by the learned trial Judge in the first trial, it will be noticed, was significantly less detailed, and covered fewer of the issues catalogued by Sully J, than the Longman direction in R v BWT (supra), which was found by the Court of Criminal Appeal not to be adequate. 148 Further, having regard to the age of the complainant, when the events which he was describing occurred, it was important that his Honour warn the jury of the fragility of recollection in one so young. 149 The directions in the second trial were more detailed. However, I believe that they also plainly failed to conform to the requirements of Longman, for many of the same reasons. 150 Although, in the context of the warning, it would have been appropriate to refer to matters which may be regarded as confirmation of the complainant's account (including the striking similarity of the evidence given by the other boy), that did not, in my view, remove the need for such warning. The disadvantage to the appellant arising from delay remained, whether there was corroboration or not. The jury needed to be told about it, so that its assessment of the complainant's evidence (and that of the other boy) could be made with an appreciation of the appellant's position. 151 In both trials, therefore, I believe that error has been established.

Rule 4 and the Proviso 152 No objection was taken at either trial to the directions which are now impugned. Rule 4 therefore applies. 153 In R v GPP [2001] NSWCCA 493, Heydon JA (with whom Wood CJ at CL and Carruthers AJ agreed) made the following remarks on the availability of Rule 4, in the context of a failure to conform with Longman: (para 60) 60. The Crown's first point in answer was that counsel for the appellant did not seek at the trial the direction which it is now said should have been given, and that leave is required under r4 of the Criminal Appeal Rules to advance these grounds. Counsel's failure is regrettable, but if the appellant's criticisms are sound, they are, in the light of the High Court authorities, sufficiently important for leave to be given, as it customarily is by this Court in cases concerning Longman directions. 154 In the same case, Heydon JA made the following comment upon the proviso: (para 111) 111. The application of the proviso depends on the error not being one which is so fundamental as to have caused the trial to miscarry so far as hardly to be a trial at all, and on the conviction being 'inevitable', so that the appellant can be said not to have lost a chance which was fairly open to him to have been acquitted. See Glennon v R (1994) 119 ALR 706 at 710-712. This error is not fundamental in the sense just described. Attention must therefore be directed to the strength of the Crown case. 155 Here, unquestionably, the Crown case against the appellant in both trials was strong. However, I do not believe in either case it can be said that conviction was inevitable (Doggett v The Queen (supra) Gaudron and Callinan (para 55) (182 ALR 1 at 12); Kirby J (para 151) (182 ALR 1 at 33)). Again, the comment of Heydon JA in R v GPP (supra) is apposite: (para 112) 112. The appellant's submission that the Crown case was weak is baseless. But once it is assumed, as it must be, that the prejudicial effect of delay on the forensic position of the appellant was not obvious to the jury, unless it had been explained to them, the warning which was not given could have caused them to experience a reasonable doubt. 156 Therefore, I believe the appropriate order is as follows: 1. The appeal in respect of each trial is upheld. 2. The conviction recorded as a result of each trial is quashed and the sentences set aside. 3. There should in each case be a new trial. **********

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.