REGINA v MARKULESKI

SPIGELMAN CJ: Tomislav Markuleski (“the appellant” or “TM”) appeals against his conviction in the District Court on 23 August 2000 in respect of five offences of a sexual nature against B P (“the complainant”). The appellant was arraigned in relation to six offences against the complainant alleged to have occurred between 29 January 1981 (count 1) and 4 February 1982 (count 6). There were four counts pursuant to s76 of the Crimes Act 1900 alleging indecent assaults of a person under 16 years (counts 1-4) and two counts pursuant to s61D(1), alleging sexual intercourse without consent with a person under 16 years (counts 5-6).

The case, as is usual in these situations, was in large measure one of word against word, i.e. that of the complainant against that of the appellant. The verdicts of the jury were guilty in relation to the first five counts and not guilty in relation to the sixth count. The course of the trial and the issues on the appeal are set out in the judgment of Simpson J, which I have read in draft. Unreasonable Verdicts

The first ground of appeal is that the verdicts of guilty in relation to counts 1-5 are “unreasonable and cannot be supported” in terms of s6 of the Criminal Appeal Act 1912, in the light of the not guilty verdict in relation to count 6. Reliance was placed on the High Court’s decision in Jones v the Queen (1997) 191 CLR 439.

The test of whether a jury verdict is “unreasonable or cannot be supported” within the meaning of s6 of the Criminal Appeal Act 1912 is as established by the joint judgment of the High Court in M v The Queen (1984) 181 CLR 487 at 494-495: “In most cases a doubt experienced by an appellate court will be a doubt which a jury ought also to have experienced. It is only where a jury’s advantage in seeing and hearing the evidence is capable of resolving a doubt experienced by a court of criminal appeal that the court may conclude that no miscarriage of justice occurred. That is to say, where the evidence lacks credibility for reasons which are not explained by the manner in which it was given, a reasonable doubt experienced by the court is a doubt which a reasonable jury ought to have experienced. If the evidence, upon the record itself, contains discrepancies, displayed inadequacies, is tainted or otherwise lacks probative force in such a way as to lead the court of criminal appeal to conclude that, even making full allowance for the advantages enjoyed by the jury, there is a significant possibility that an innocent person has been convicted, then the court is bound to act and to set aside a verdict based upon that evidence. In doing so, the court is not substituting trial by a court of appeal for trial by jury, for the ultimate question must always be whether the court thinks that upon the whole of the evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty.” [references omitted]

The application of this test in a case of alleged inconsistency between a verdict or verdicts of acquittal and of conviction received an authoritative exposition in the joint judgment of Gaudron, McHugh and Gummow JJ in Jones v The Queen esp at 453-455. Their Honours quoted extensively from the judgment in M .

The general principles with respect to inconsistent verdicts had been set out in authorities before Jones , particularly MacKenzie v The Queen (1996) 190 CLR 348 at 366-368 per Gaudron, Gummow and Kirby JJ. The threshold proposition, relevant for present purpose, enunciated by their Honours in MacKenzie at 366 was: “Where, as is ordinarily the case, the inconsistency arises in the jury verdicts upon different counts of the originating process in a criminal trial, the test is one of logic and reasonableness.”

The reasoning of the joint judgment in Jones , where there was nothing illogical about the divergent verdicts, represents a particular application of this test of “reasonableness”.

The reasoning in MacKenzie at 367 is pertinent to the present case: “… the respect for the function which the law assigns to juries (and the general satisfaction with their performance) have led courts to express repeatedly, in the context both of criminal and civil trials, reluctance to accept a submission that verdicts are inconsistent in the relevant sense. Thus, if there is a proper way by which the appellate court may reconcile the verdicts, allowing it to conclude that the jury performed their functions as required, that conclusion will generally be accepted. If there is some evidence to support the verdict said to be inconsistent, it is not the role of the appellate court, upon this ground, to substitute its opinion of the facts for one which was open to the jury. In a criminal appeal, the view may be taken that the jury simply followed the judge’s instruction to consider separately the case presented by the prosecution in respect of each count and to apply to each count the requirement that all of the ingredients must be proved beyond reasonable doubt.” [references omitted]

The respect which the Court pays to the constitutional role of the jury was emphasised in M at 493, in a passage referred to with approval in Jones at 451: “… the question which the court must ask itself is whether it thinks that upon the whole of the evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty. But in answering that question the court must not disregard or discount either the consideration that the jury is the body entrusted with the primary responsibility of determining guilt or innocence, or the consideration that the jury has had the benefit of having seen and heard the witnesses. On the contrary, the court must pay full regard to these considerations.” [references omitted]

It is against this background, that the test for an unreasonable verdict - whether it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty - must be applied to the whole of the facts and circumstances of a particular case.

I do not understand there to be any relevant difference between a ‘no reasonable jury test’ ( MacKenzie at 366 quoting Lord Devlin) and a ‘not open to the jury’ test ( Jones at 450-452 adopting M ).

In Jones, especially at 453-455, passages of which I will set out below, the joint judgment referred to three matters in detail. First, the effect on the credibility of the complainant of the acquittal on one count. Second, the absence of corroboration. Third, the difficulties occasioned to the defence by the unexplained long delay in making a complaint. The joint judgment concluded at 455: “When the credibility factor is combined with the uncorroborated nature of the complaints and the effect of the lengthy and unexplained delay in the making of the complaints, the convictions on the first and third counts can only be regarded as unsafe and unsatisfactory.”

As to the first matter, their Honours analysed the effect of the acquittal on one count upon the credibility of the complainant. I will refer to this passage in detail below.

As to the second matter the joint judgment referred at 453 to: “the lack of any corroborative evidence, in particular, the absence of any medical evidence” [reference omitted] as a factor which: “made it necessary for a reasonable jury to scrutinise the complainant’s evidence with considerable care”.

As to the third matter, their Honours said at 454: “The unexplained absence of a complaint for four years is a matter that made it essential for a reasonable jury to scrutinise the complainant’s evidence with great care.” And continued: “Furthermore, in some cases - and we thought that this was one - the delay in making a complaint may be so long that it hampers an accused person’s right to defend him or herself.” The joint judgment at 454-455 referred to the specific difficulties which delay had posed for the defence in the particular circumstances of the case.

Accordingly, it was the combined operation of three factors - credibility, absence of corroboration and the effect of a lengthy and unexplained delay - which led the court to conclude that it was not open to the jury to convict. Two of these factors - credibility and effect of delay - involve matters of fact and degree the significance of which turned on the particular circumstances of the case.

A noteworthy feature of Jones was the fact that the trial judge did not give a warning to the jury about the adverse effects of delay upon the ability of the accused to defend himself. There was no warning in accordance with the decision in Longman v The Queen (1987) 168 CLR 79 (reaffirmed in Crampton v The Queen (2000) 75 ALJR 133; [2000] HCA 60).

In Jones the Court of Criminal Appeal rejected leave under r 4 of the Criminal Appeal Rules to raise this issue. Leave was required because no application had been made to the trial judge after his summing-up for a relevant redirection. See R v VRJ (NSWCCA, 22 November 1996, unreported).

In the High Court the fourth member of the majority, Brennan CJ, at 445-446, expressly referred to the absence of a Longman warning as a relevant factor in his Honour’s conclusion that the verdict on one count was unsafe and unsatisfactory. Although the joint judgment does not explicitly refer to the absence of a Longman warning, the significant emphasis that their Honours gave to the adverse effects of delay in concluding that the verdicts were unsafe, may also have been affected by the fact that the jury was not instructed about the adverse effects of delay.

The absence of a Longman warning in Jones , was not raised directly in the High Court. It appears to have been subsumed (expressly in the judgment of Brennan CJ and implicitly in the joint judgment) under the “unsafe and unsatisfactory” ground. The apparent inconsistency of the verdicts in Jones may have appeared to have been of lesser force if a full Longman warning had been given. The element of delay, which the joint judgment mentioned as one of the factors which led their Honours to conclude that it was not open to the jury to convict, would have appeared in a quite different light.

This aspect of Jones has not received appropriate emphasis in subsequent cases. In the case law since Jones , as in the present case, attention has focused on the consideration of the “credibility factor” in the joint judgment.

Australian jurisprudence on inconsistent verdicts has developed a much higher degree of intolerance to inconsistency than appears to be the case in some other jurisdictions.

In England, save in exceptional cases, “logical inconsistency is an essential prerequisite for success on this ground”. ( R v W, Court of Appeal (Criminal Division) 30 March 1999, unreported, per Lord Bingham of Cornhill CJ). The fact that different counts all depended on the uncorroborated evidence of the same witness does not render different verdicts inconsistent. (See the following cases decided in the Court of Appeal (Criminal Division): R v Cilgram [1994] Crim LR 861; R v Warner (17 February 1997, unreported); R v Bell (15 May 1997, unreported); R v Clarke & Fletcher (30 July 1997, unreported); R v Britton (5 October 2000, unreported)). In R v Guest (23 January 1998) partially reported as R v G [1998] Crim LR 483, the court was referred to Jones but concluded: “… the mere fact that credibility of the complainant is in issue and her evidence is acted upon in some counts and not in others does not constitute a logical inconsistency.”

New Zealand courts also apply a “no reasonable jury” test. (See R v Irvine [1976] 1 NZLR 96 at 99; R v Kino and Mete [1997] 3 NZLR 24 at 27). The position appears to be broadly similar to the approach in Australia. (See R v O (No. 2) [1999] 1 NZLR 326; R v H [2000] 2 NZLR 581).

Canadian courts are prepared to inquire whether the error was in the acquittal rather than in the conviction ( Koury v The Queen (1964) 43 DLR (2d) 637 at 652). A diversity of approach is apparent in R v R (D) (1996) 136 DLR (4th) 525 which is of limited assistance as it was an appeal following a trial by judge alone.

In the United States most jurisdictions do not intervene even with logically inconsistent verdicts (see E. L. Muller “The Hobgoblin of Little Minds? Our Foolish Law of Inconsistent Verdicts” (1998) 111 Harvard Law Review 77). Application of the Reasoning in Jones

Since the High Court decided Jones there have been numerous decisions by courts of criminal appeal where a complete acquittal has been obtained by an appellant who had been found guilty of some, but not all, sexual assault counts. These include: New South Wales R v Crisologo (1997) 99 A Crim R 178 (New trial ordered) R v AM (NSWCCA, 2 April 1998, unreported) R v Wells (NSWCCA, 17 June 1998, unreported) R v DSA [1999] NSWCCA 35 R v Cox [1999] NSWCCA 62 R v James [1999] NSWCCA 191 R v Meola [1999] NSWCCA 388 R v RAT (2000) 111 A Crim R 360 R v Robinson (2000) 111 A Crim R 388 (New trial ordered) R v Barnett [2000] NSWCCA 283 R v AT [2000] NSWCCA 342 R v Carbone [2000] NSWCCA 387 R v Whitehead [2000] NSWCCA 400 R v GWB [2000] NSWCCA 410 R v Mills [2001] NSWCCA 48 R v S [2001] NSWCCA 204 R v SAW [2001] NSWCCA 222 R v GWM [2001] NSWCCA 267 Victoria R v JGVR [2001] VSCA 8 Queensland R v Watson (1997) 95 A Crim R 98 R v Russell (QCA, 24 March 1999, unreported) R v P [2000] 2 Qd R 401 R v D [2000] QCA 417 R v Kenward [2000] QCA 482 R v Ritchie [2001] QCA 142 Western Australia Wheeler v The Queen (WASCA, 20 April 1998, unreported) M (1998) 104 A Crim R 154 Eastough v The Queen (WASCA, 12 March 1998, unreported) Northern Territory Lajciak v The Queen [1999] NTSC 82

As can be seen, the application of the reasoning in Jones has had a substantial impact on the administration of criminal justice, particularly in this State.

Cases in which an appeal on the grounds of inconsistent verdicts has not been allowed include: New South Wales R v DJK (1997) 96 A Crim R 443 R v JJT (NSWCCA, 3 December 1997, unreported) R v Kiskarpati (NSWCCA, 4 November 1998, unreported) R v W (1999) 109 A Crim R 51 R v AG & RJM [2000] NSWCCA 86 R v Steenson [2000] NSWCCA 143 R v ACK [2000] NSWCCA 180 R v OGD (No. 2) [2000] NSWCCA 404 R v ARD [2000] NSWCCA 443 R v Henman [2001] NSWCCA 4 R v MFA [2001] NSWCCA 71 R v PLV [2001] NSWCCA 282 Victoria R v KET [1998] VSCA 73 R v Hilsley [1998] VSCA 143 R v J (No 2) [1998] 3 VR 602 Queensland R v Knuth (QCA, 23 June 1998, unreported) R v Petterwood (QCA, 11 August 1998, unreported) R v McNamara (QCA, 1 December 1998, unreported) R v Maddox (QCA, 4 December 1998, unreported) R v R [2000] QCA 27 R v C [2000] QCA 385 R v A [2000] QCA 520 Western Australia Ellis v The Queen (WASCA, 26 September 1997, unreported) King, Bull and Marotta v The Queen (WASCA, 19 January 1998, unreported) Reppas v The Queen (WASCA, 9 December 1998, unreported) Miles v The Queen (WASCA, 21 December 1998, unreported) RM v The Queen (WASCA, 23 February 1999, unreported) Asplin v The Queen (WASCA, 26 August 1999, unreported) Strickland v The Queen [2000] WASCA 68 Wimbridge v The Queen [2000] WASCA 117 Jaensch v The Queen [2000] WASCA 212 Balaam v The Queen [2000] WASCA 292 Miles v The Queen [2000] WASCA 364 Spiteri v The Queen [2001] WASCA 82 Blight v The Queen [2001] WASCA 122 South Australia R v T (1999) 74 SASR 486 R v Tee [1999] SASC 352 R v KWG [2000] SASC 398 Tasmania Duniam v R (TASSC, 22 September 1997, unreported) Williams v R [2000] TASSC 182

Differences in approach are apparent within this Court and between judgments of this Court and of courts of criminal appeal in other states. In these circumstances, the Court has convened a bench of five. The Duty of the Jury to Consider Charges Separately

Nothing in Jones casts any doubt on the appropriateness of a jury accepting a witness’ evidence in one respect, whilst retaining a reasonable doubt about the commission of events about which that same witness is the only substantive witness.

Indeed, in a case handed down by the High Court exactly one week after Jones ( KBT v The Queen (1997) 191 CLR 417) Brennan CJ, Toohey, Gaudron and Gummow JJ at 424 emphasised the importance of the instruction traditionally given in this respect: “As the trial judge correctly instructed the jury in his summing up, it was open to the jury to accept some parts of M’s evidence and to reject others.”

The appropriateness of the direction traditionally given has also recently been confirmed by McHugh J in KRM v The Queen (2001) 75 ALJR 550; [2001] HCA 11 where his Honour said: “[36] It has become the standard practice in cases where there are multiple counts, however, for the judge to direct the jury that they must consider each count separately and to consider it only by reference to the evidence that applies to it (a “separate consideration warning”). The universal giving of a separate consideration warning and the omission of a universal propensity warning indicates that the giving of a separate consideration warning is ordinarily sufficient to avoid miscarriages of justice in cases such as the present. This indication is confirmed by the many cases where juries acquit accused persons of some charges and convict them of others where the presentment contains multiple counts involving the same or similar offences. Indeed, so freely do juries acquit of some charges and convict of others on presentments with multiple counts that appellate courts often hear arguments that there is such an inconsistency in the verdicts that the convictions are unsafe and must be set aside [60] … [60] cf Jones v The Queen (1997) 191 CLR 439.”

In the common case of multiple sexual assaults against a single complainant, often over a period of time, juries frequently acquit on some charges and convict on others. The issue raised by Jones is to determine when an acquittal so affects the credibility of the complainant that, in combination with other factors, a conviction was not open to the jury. A court of criminal appeal must perform this task whilst acknowledging the role of the jury as emphasised in M, MacKenzie and Jones quoted above. The Centrality of the Facts of a Particular Case

Several aspects of the reasoning in Jones pertinent to the issue of credibility of the complainant in that case, often arise in other cases. These include: · Absence of corroboration in a word against word case. · Delay in making complaint, · Lack of apparent difference in the quality of the complainant’s evidence amongst the different incidents. · Availability of contradictory evidence with respect to some, but not all, incidents.

Although the specific aspects of a case may be characterised in this way - including in the specific language, quoted below, of the reasoning in Jones - each such matter raises questions of fact and degree which may affect the ultimate assessment of the effect upon the credibility of the complainant in the particular case before the Court when the complainant’s evidence is not accepted to the criminal standard on some counts. The fact that the High Court found in Jones that the effect on credibility was such that, together with other factors, it was not open to the jury to convict, does not determine the outcome of other cases. The variation in fact and degree will differ from case to case.

The proposition that these cases turn on their particular facts is confirmed by High Court authority, both before and after Jones .

In MacKenzie the joint judgment relevantly concluded at 368: “It is only where the inconsistency rises to the point that the appellate court considers that intervention is necessarily required to prevent a possible injustice that the relevant conviction will be set aside. It is impossible to state hard and fast rules. ‘It all depends upon the facts of the case .’” [Emphasis added; reference omitted]

In Osland v The Queen (1998) 197 CLR 316, inconsistency of verdicts was raised in the context of allegations that the appellant Osland and her son had murdered the appellant’s husband, the son’s stepfather. At trial the mother was convicted, but the jury was unable to reach a verdict with respect to the son. The son was later tried again and acquitted. The assertion that the verdicts were inconsistent was raised for the first time in the High Court.

The appeal was dismissed by majority. Each of the majority judgments gave careful attention to the particular facts and issues raised in the course of the proceedings. I do not understand either Kirby J or Callinan J to disagree with the general thrust of the following reasoning of McHugh J: “[120] When there is no legal or technical inconsistency in the verdicts, a conviction can only be set aside where the inconsistency in the verdicts demonstrates that no reasonable jury who had applied their minds to the evidence could have arrived at the two different verdicts. In such a case the Court of Criminal Appeal sets aside the conviction because it is unsafe or unsatisfactory. [121] In determining whether the inconsistency points to an unsatisfactory conviction, the appellate court must consider the evidence, the issues, and the directions which the jury were given. … An examination of the directions, issues and evidence may confirm that apparently inconsistent verdicts are in fact inconsistent and demonstrate that the conviction is unsafe. Just as frequently, however, examination of the issues, evidence and directions may show that apparently inconsistent verdicts are not inconsistent and that there has been no failure in the reasoning process of the jury.” [references omitted]

The central significance of the facts of a particular case is also emphasised in the High Court judgment in Farrell v The Queen (1998) 194 CLR 286. The appellant had been charged with eight counts of rape, two counts of aggravated sexual assault and two counts of assault. He was found guilty of one count of assault, one count of aggravated sexual assault (by majority) and four counts of rape (three by majority). The jury returned verdicts of not guilty on five counts and on one count there was a verdict of not guilty by direction. All of the offences had occurred in the course of one night on which the complainant said that he had been sexually assaulted and beaten by two men. Two of the five counts of which the appellant was acquitted were those in which the principal offender was alleged to be the other person, with the appellant being charged with aiding and abetting that person.

Hayne J noted at [64] that there was very little evidence of the appellant encouraging the other person by his alleged presence. His Honour concluded at [65]: “In those circumstances I attach no weight to the different verdicts given by the jury to the various counts charged. The verdicts are entirely consistent with the jury not reaching the requisite degree of satisfaction that the appellant had aided and abetted Frank in his assaults on the complainant.” McHugh and Gaudron JJ agreed with Hayne J.

Both Kirby J at [25] and Callinan J at [99] referred to the differentiation that Cox CJ had made in the Court of Criminal Appeal. Cox CJ had said ( Farrell v R (TASSC, 7 June 1996, unreported)): “[The jury] did not, in fact, convict on any counts proof of which required a finding of the commission of a crime by the third man. … There was no corroborative evidence in respect of such counts - no corroborative evidence even of the existence, let alone presence, of that person; but as there was, in my view, significant corroborative evidence in the form of the tie which linked the appellant to the complainant not only earlier in the evening but up until shortly before the complainant’s arrival at his brother’s home with his hands bound, it seems likely that the jury were fully alive to the deficiencies in the complainant’s credibility and were only prepared to convict where his evidence was in fact corroborated.”

Callinan J said at [99]: “In the Court of Criminal Appeal, Cox CJ explained this differentiation in the verdicts. The evidence concerning the distinctive tie was logically corroborative of much of the complainant’s evidence and capable clearly of linking the appellant with the commission of the offences of which he was charged. There was no similar corroboration of the offences involving the second man. The position may simply have been that the jury were left in a state of doubt about those offences. It cannot be said that there is no rational basis for the respective verdicts of guilty and not guilty.”

Only two of the five counts on which the appellant was found to be not guilty were aiding and abetting counts. He was also found not guilty of one of the six counts of anal penetration, one count of fellatio and one count of assault (see [98] in the judgment of Callinan J). It does not appear that the presence or absence of the third man was relevant to these counts. Nevertheless, no fatal inconsistency was found to exist on the facts of this case, although no separate consideration was given to the three counts not involving the third man for which an acquittal was found.

In my opinion, Osland and Farrell support the proposition that the reasoning in Jones is directed to the facts of that case. The Reasoning in Jones

There are three key passages in the reasoning of the joint judgment in Jones on the issue of credibility. (i) “The jury’s finding of not guilty on the second count damaged the credibility of the complainant with respect to all counts in the indictment. Implicit in the appellant’s acquittal on the second count was a rejection of the complainant’s account of the events which were said to give rise to that count.” (453) (ii) “It is difficult then to see how it was open to the jury to be convinced beyond a reasonable doubt of the guilt of the appellant with respect to the first and third counts. There is nothing in the complainant’s evidence or the surrounding circumstances which gives any ground for supposing that her evidence was more reliable in relation to those counts than it was in relation to the second count.” (453) (iii) “In our opinion, the proper application of the test formulated by the majority of this court in M required the convictions of the appellant to be set aside on the basis that they were unsafe and unsatisfactory. Given the jury’s findings on the second count, it was not open to them, on the whole of the evidence, to be satisfied beyond reasonable doubt of the guilt of the appellant on the first and third counts. Once the jury found that the evidence of the complainant with respect to the second count lacked sufficient cogency to convict, the Crown case on the first and third counts wore a different complexion. For it meant that, when her evidence could be set against other reliable evidence, it failed to carry sufficient conviction to reach the criminal standard of proof. As we have already said, nothing in the complainant’s evidence gave any ground for thinking that the quality of her evidence was higher in respect of the first and third count than it was in respect to the second count.” (455)

These considerations operated in a, frequently occurring, context of delay in complaint and absence of corroboration which, as the joint judgment emphasised, required the jury “to scrutinise the complainant’s evidence with considerable care” (453.9).

In each of passage (ii) and (iii), the joint judgment notes that there was no reason to believe that the complainant’s evidence with respect to those counts on which a conviction was recorded was either “more reliable” (453.8) or of a “higher quality” (455.8). However, I do not understand their Honours to be propounding a proposition that unless a finding of that character can be made, then in a word against word case where the complainant’s version is uncorroborated, the only conclusion is that it was not open to the jury to convict. This appears from the addition of the words “ or the surrounding circumstances” [emphasis added] in passage (ii).

In passage (iii), their Honours said that the count on which an acquittal was entered was the only count on which, given the delay, the complainant’s evidence “could be set against other reliable evidence” (455.7). Again, I do not understand that their Honours were propounding a proposition that in any word against word case, where such a characterisation can be made, the only conclusion is that it was not open to the jury to convict.

Both of these considerations are, of course, important and often critical, but their significance gives rise to questions of fact and degree which will vary from case to case. The reasoning has not always been treated in this way. Application of the Reasoning in Jones

The reasoning in Jones appears to have been treated as determinative of any analogous case by Dunford J in R v RAT [2000] 111 A Crim R 360 when his Honour said: “[44] The fact remains that all counts depended on the evidence of the complainant being accepted as true beyond reasonable doubt and on the only two counts where her evidence could be set against other apparently or potentially reliable evidence, it failed to carry sufficient conviction to reach the criminal standard of proof in the eyes of the jury. There is nothing in her evidence or the surrounding circumstances which gives any ground for supposing that her evidence was more reliable in relation to the first three counts where it could not be tested than it was in relation to the other counts where it could be, and was, tested. On the two counts where on the complainant’s evidence, corroboration was available, such corroboration was not forthcoming. The very matters which justify a finding that the verdicts are not inconsistent in the sense of being a compromise are matters which call into question the complainant’s credibility in respect of the counts on which the appellant was acquitted. [45] In those circumstances, notwithstanding the fact that her evidence as transcribed carries conviction, that it does not, in my opinion, contain discrepancies or inconsistencies such as were apparent in M and the trial judge who like the jury had the opportunity of seeing the complainant and other witnesses was satisfied she was telling the truth, I consider that in accordance with Jones the convictions on counts 1-3 must be regarded as unreasonable and cannot be supported, and must be set aside and judgments of acquittal entered.” [Emphasis added]

The matters emphasised in par [44] of Dunford J’s judgment paraphrase the terminology in the majority judgment in Jones which his Honour had earlier set out. His Honour’s recitation of countervailing facts in par [45] are, in my opinion, facts and matters relevant to a conclusion in that, or any analogous case, that the result in Jones should not apply on the facts in that case. Nevertheless his Honour considered that “in accordance with Jones ”, the court should set aside the convictions in the case before the court. In my opinion, it is wrong to approach the authority of Jones on the basis that whenever facts of a similar character are found to those set out in the passages of the joint judgment to which I have referred above, the finding must be the same as in Jones . Whether or not such a conclusion is appropriate depends on all of the circumstances of the case.

Dunford J went on to reiterate his understanding of the effect of Jones as a precedent in the context of formulating a direction which he concluded ought to be given in such cases. His Honour said: “[46] If I am correct in my understanding of the effect of Jones, then I believe consideration needs to be given to the directions regularly given to juries in such trials. Juries are constantly told that they should consider each count in the indictment separately and also that they may accept part of a witness’ evidence and not accept other parts of the evidence of that same witness. But Jones appears to establish a qualification to these directions to the effect that in sexual cases, where the only direct evidence of the commission of the offences is that of the complainant, if they are for any reason not satisfied beyond reasonable doubt that the complainant is telling the truth in relation to one count, it is not open to them to be satisfied to the criminal standard that she (or he) is telling the truth in relation to any other count; and therefore in such cases (that is, where there is no independent evidence of the commission of the offences) if they find an accused not guilty on one count, they must also find him not guilty on all counts. They should also be warned specifically that if they return different verdicts where there are no distinguishing features in the evidence, such verdicts are liable to be regarded as a compromise and the guilty verdicts set aside.”

His Honour’s statement that, in the identified circumstances, the jury “must … find” the accused “not guilty on all counts” elevates the conclusion in Jones to a proposition of law. In my opinion, this approach is wrong. The appropriateness of the suggested direction will be discussed below.

In R v Barnett [2000] NSWCCA 283, Dunford J applied the approach he had adopted in RAT . In Barnett the counts on which the appellant was convicted were distinguished by the fact that there was independent evidence of opportunity to commit the offences. His Honour found that, applying MacKenzie , the different verdicts were not “an affront to logic and commonsense” but, nevertheless, applying Jones: “[28] The independent evidence of opportunity of them being alone … did not render her evidence more credible on those counts than on the others.”

In R v Carbone [2000] NSWCCA 387, Wood CJ at CL referred to the decisions in this Court applying Jones and to Dunford J’s analysis of the cases in RAT. Wood CJ at CL added: “[56] … In summary those decisions give recognition to the circumstance that,where in sexual assault cases with multiple counts involving a single complainant and a single accused, the only direct evidence of the commission of the offences is that of the complainant, then acquittal on any one of the charges will in the absence of an acceptable explanation, normally require acquittal on the remainder. [57] This follows from the circumstance that, in such a case, the rejection of the complainant’s evidence will involve a diminution in his or her overall credibility which cannot be overcome, in the absence of some circumstance that might restore that credibility for the other counts. This requires identification of some fact or surrounding circumstance that would provide a rational basis for accepting the complainant’s evidence upon the other counts, ie something which would justify the conclusion that the evidence, on those other counts, was more reliable, or stronger, than it was for the count upon which it was rejected.”

Similarly, in R v Dixon [2001] NSWCCA 39, Whealy J said at [52]: “In R v RAT (2000) NSWCCA 77 Dunford J (with whom Barr J agreed) at (46) held that Jones v The Queen (1997) 191 CLR 439 appears to have established that in sexual assault cases where the only direct evidence of the commission of the offences is that of the complainant, if the jury is for any reason not satisfied beyond reasonable doubt that the complainant is telling the truth in relation to one count, it is not open to them to be satisfied to the criminal standard that the complainant is telling the truth in relation to any other count. Therefore, in such cases, if the jury finds an accused not guilty on one count it must also find him not guilty on all counts.”

To similar effect is the conclusion of Sully J in R v S [2001] NSWCCA 204. In that case the appellant had been convicted on three counts and acquitted on three counts. With respect to the latter the complainant’s evidence about the surrounding circumstances of the alleged incidents had been successfully challenged. Sully J, with whom Dowd J agreed, referred to Jones at 453 and added at [14]: “ … whilever that reasoning remains on foot with the authority of the High Court of Australia then this applicant would, given the facts of the present case seem to me to entail necessarily that the convictions on counts 1, 2 and 3 cannot stand.”

The approach to the reasoning in Jones reflected in RAT, Barnett, Carbone, Dixon and S has not been universally applied in this Court. If it had been adhered to, it is likely that a number of decisions would have been different (e.g. R v JJT (NSWCCA, 3 December 1997, unreported); R v Kiskarpati ( NSWCCA, 4 November 1998, unreported); R v W (1999) 109 A Crim R 51; R v AG & RJM [2000] NSWCCA 86; R v ACK [2000] NSWCCA 180; R v AT [2000] NSWCCA 342 (as to the first count at [79]); R v ARD [2000] NSWCCA 443; R v MFA [2001] NSWCCA 71).

Furthermore, the approach does not appear to have been adopted by courts of criminal appeal in other States. (See e.g. R v J (No 2) [1998] 3 VR 602 esp at 628; R v KET [1998] VSCA 73 at [29]; R v Hilsley [1998] VSCA 143 at [60] and [78]; R v Maddox (QCA, 4 December 1998, unreported); R v Petterwood (QCA ,11 August 1998, unreported); R v Knuth (QCA, 23 June 1998, unreported); R v R [2000] QCA 27 at [26]-[28]; R v C [2000] QCA 385 at [47]; R v KWG [2000] SASC 398 at [65]-[67], [77]-[83]; RM v The Queen (WASCA, 23 February 1999, unreported); Ellis v The Queen (WASCA, 26 September 1997, unreported); Williams v R [2000] TASSC 182 at [6]; Asplin v The Queen, (WASCA, 26 August 1999, unreported) at [26] and [39]; Balaam v The Queen [2000] WASCA 292 at [17]; Spiteri v The Queen [2001] WASCA 82 at [6]).

There are, however, cases which appear to be consistent with the approach in the line of New South Wales authorities to which I have referred (see e.g. R v Kenward [2000] QCA 482 at [6], [8] and [149]-[150]; R v Ritchie [2001] QCA 142).

In these circumstances it is appropriate that prior authority in this Court should be reviewed by a bench of five.

The reasoning in RAT, Barnett, Carbone, Dixon and S suggests that Jones established the proposition that in a pure word against word case a mixture of acquittals and convictions cannot be sustained unless the Court can detect a relevant difference in the quality of the complainant’s evidence.

In my opinion this is not what the High Court intended. Whether or not the failure of the jury to accept the complainant’s version in one respect ought to have led to the jury to have a reasonable doubt with respect to other matters, must depend on the full range of relevant circumstances. The High Court held in Jones that on the facts of that case the acquittal necessarily undermined the credibility of the complainant’s evidence about other alleged incidents. By reason of the wide range of matters of fact and degree that must be considered in making a credibility finding, that conclusion does not, in my opinion, follow in every such case unless the Court is positively satisfied that there is some relevant difference in the quality of the complainant’s evidence.

I repeat, with my emphasis, a sentence from the joint judgment in Jones at 453: “There is nothing in the complainant’s evidence or the surrounding circumstances which gives any ground for supposing that her evidence was more reliable in relation to those counts than it was in relation to the second count.” The reference to “the surrounding circumstances” requires an appellate court to consider all the facts of the case.

I agree with the reasoning of Winneke P in R v KET [1998] VSCA 73: “[28] … Jones’s case, in my view, is a case which has to be looked at in the light of its own particular facts, as indeed all cases where this issue raised must be. In Jones’s case three discrete acts of sexual penetration were alleged and the jury convicted the accused on two and acquitted on one, in circumstances where the acquittal, as the judgments made clear, must necessarily on the facts of that case have reflected upon the credibility of the complainant’s evidence on the other two counts. Furthermore, it would seem from the judgments in Jones’ case that special emphasis was laid on the prejudice flowing to the accused from the delay in complaint because it was apparent from the special facts of the case that there was a real risk that the delay had deprived the accused of what the majority judges had called ‘a cast iron alibi’. [29] However, in my view, it would be wrong to draw from the decision of Jones’ case the general proposition that, in cases where multiple sexual offences are alleged involving the one complainant, the jury’s acquittal on some of those counts should compel an appellate court to conclude that the jury must necessarily have regarded the complainant generally as an untruthful witness or that her credibility was impacted upon in respect of the counts upon which they had convicted.” (See also R v KWG [2000] SASC 398 at [79] per Martin J).

There have, of course, been cases since Jones in which the Court was able to detect from the transcript that there were relevant differences in the way the complainant gave evidence on the matters which led to an acquittal when compared to those which led to a conviction (see e.g. R v RJB [1999] VSCA 188 at [26]; R v Tee [1999] SASC 352 at [70]; R v A [2000] QCA 520 at [23] and [106]; Jaensch v The Queen [2000] WASCA 212 at [37]; Miles v The Queen [2000] WASCA 364 at [66]; R v OGD (No. 2) [2000] NSWCCA 404 at [126]; Strickland v The Queen [2000] WASCA 68 at [65]-[70], [74]; Duniam v R (TASSC, 22 September 1997, unreported).

The absence of any relevant difference in the complainant’s evidence was emphasised in Jones , as it had been in M (at 494 quoted at para 4 above). It is not, however, in my opinion, the case that the reasoning in Jones requires a court of criminal appeal to quash a conviction in a pure word against word case, unless the differences between or amongst the verdicts can be explained in this way.

Indeed, the courts’ respect for the constitutional role of the jury, affirmed in Jones itself, is such that a court of criminal appeal could conclude that subtle differences in the way evidence was presented - differences not apparent from the transcript available on appeal - may explain what appear to be, at first sight, inconsistent verdicts.

In R v J (1994) 75 A Crim R 522 at 540, Brooking J referred to “the inability of the transcript to record pauses or other indications of indecision”. In R v Robinson (2000) 111 A Crim R 388 at [61] Barr J referred to the possibility that the jury may make an assessment of reliability “… by the way he gave his evidence rather than by what he said”. Adams J made similar observations in ARD supra at [201] and [203]. See also Spiteri v The Queen [2001] WASCA 82 at [34].

Furthermore, in a number of cases a court of criminal appeal has expressed its conclusion in terms of what a jury “may” have done, which suggests that some form of differentiation about the character of the complainant’s evidence was accepted by the jury, although it was not apparent to the appellate court. (See e.g. Miles v The Queen (WASCA, 21 December 1998, unreported); R v Hilsley [1998] VSCA 143 at [78]; R v C [2000] QCA 385 at [47]; R v Knuth (QCA, 23 June 1998, unreported); Balaam v The Queen [2000] WASCA 292 at [9], [14], [17]; Spiteri v The Queen [2001] WASCA 82 at [22], [32] and [37]).

The circumstances of a particular case may lead to the conclusion that a jury which has found that it had a reasonable doubt with respect to a complainant’s evidence on one count, ought to have had such a doubt with respect to other counts. However, it does not follow that that must necessarily be the case.

It is not, however, easy to formulate principles as to when a jury should conclude that a reasonable doubt on one count, notwithstanding the complainant’s evidence on that matter, means that the jury ought to have a reasonable doubt about other counts, about which the only evidence is from the complainant. When is Partial Rejection of the Complainant’s Evidence Acceptable?

The joint judgment in MacKenzie supra at 367-8 quoted with approval from the judgment of King CJ in R v Kirkman (1987) 44 SASR 591 at 593: “Sometimes juries apply in favour of an accused what might be described as their innate sense of fairness and justice in place of the strict principles of law. Sometimes it appears to a jury that although a number of counts have been alleged against an accused person, and have been technically proved, justice is sufficiently met by convicting him of less than the full number. This may not be logically justifiable in the eyes of a judge, but I think it would be idle to close our eyes to the fact that it is part and parcel of the system of administration of justice by juries. Appellate courts therefore should not be too ready to jump to the conclusion that because a verdict of guilty cannot be reconciled as a matter of strict logic with a verdict of not guilty with respect to another count, the jury acted unreasonably in arriving at the verdict of guilty.”

This particular application of the principle that an appellate court should respect the constitutional role of the jury, must be given careful consideration in cases alleging inconsistency in verdicts. There are often a number of possible explanations for divergent verdicts that do not necessarily give rise to a doubt about the complainant’s credibility on all counts. For example, where there is contradictory evidence on one count but not on others, the jury may well have taken the view that there was no point in adding yet another verdict of guilty, so that it was unnecessary to fully consider the conflicting evidence.

Considerations of this character appear to lie behind this Court’s judgment in R v Kiskarpati (NSWCCA, 4 November 1998, unreported) in which three counts of sexual intercourse without consent and one count of common assault were alleged to have occurred on the same occasion. The allegations relating to the sexual offences involved digital penetration, cunnilingus and vaginal intercourse. The appellant was convicted of the count of vaginal intercourse and acquitted of the other two sexual offences. Hidden J with whom McInerney and Ireland JJ agreed concluded: “… I think it sufficient to say that the jury may have taken a practical view of the whole encounter and decided that those two verdicts of guilty, that is, in relation to one charge of sexual intercourse without consent and in relation to the charge of common assault, were sufficient to reflect the appellant’s culpability.”

There are cases in which nothing at all appears to differentiate the complainant’s evidence, which the jury accepted beyond reasonable doubt, from the evidence which the jury did not so accept. In such a case the MacKenzie test of ‘logic and reasonableness’ is not satisfied.

Difficult issues arise when a reason for differentiating between the verdicts does exist but that reason relates to the credibility - whether in terms of veracity or reliability - of an uncorroborated complainant. The case law since Jones has frequently concluded that a jury was not obliged to have a reasonable doubt about the complainant’s credibility in general, when it did not accept her (it usually is a “her”) version of events to the criminal standard in some respect or respects. The distinction has been variously expressed: · “accepted the complainant as truthful but admitted the possibility of faulty recollection in some cases” R v J (1994) 75 A Crim R 522 at 540 referred to with approval in R v J (No 2) [1998] 3 VR 602 at 628. · distinguish a case in which the jury “rejected as unreliable the substance of the complainant’s evidence” from a case in which it was “cautious about some details” ( R v Maddox (QCA, 4 December 1998, unreported) at [1] per McPherson JA) and “doubts” about “details” (per Thomas JA at [21] and White J at [2]). · “still accept the basic story at less than the full alleged detail” ( R v Russell (QCA, 24 March 1999, unreported)). · “generally truthful, even if she might have been exaggerating in this part of her evidence” ( R v ACK [2000] NSWCCA 180 at [53]). · whether the jury found the complainant’s evidence “dishonest rather than unpersuasive in the circumstances” ( R v C [2000] QCA 385 at [47]). · “the jury did not disbelieve the complainant but they thought her evidence lacked the requisite particularity as to time, place or circumstance to justify conviction” ( Duniam v R (TASSC, 22 September 1997, unreported)).

In a number of other cases, courts of criminal appeal have concluded that in the circumstances of the case, an acquittal contrary to the complainant’s evidence did not necessarily mean that the complainant’s general credit was undermined (see e.g. R v KET supra at [29]; R v Hilsley [1998] VCSA 143 at [78]; R v Maddox supra at [1]; R v R [2000] QCA 27 at [28]; R v AT [2000] NSWCCA 342 at [18]; R v AG & RJM [2000] NSWCCA 86 at [51]; Williams v R [2000] TASSC 182 at [4]; R v McNamara (QCA, 1 December 1998, unreported)).

It is usual in such cases to give a strong warning about the dangers of acting on the uncorroborated evidence of the complainant. Where the jury accepts and implements such a direction, it does not necessarily mean that the jury ought to have had a doubt about the complainant’s credibility generally.

In a number of cases the presence or absence of corroboration has been accepted as a relevant point of differentiation. The jury convicted in cases where there was corroboration, but not otherwise (see e.g. King, Bull and Marotta v The Queen (WASCA, 19 January 1998, unreported) (an appeal to the High Court was successful on other grounds - Bull v The Queen (2000) 201 CLR 443); R v Petterwood (QCA, 11 August 1998, unreported); R v Hilsley [1998] VSCA 43 at [60] and [78]; R v KCW [1999] NSWCCA 112 at [95]-[106]; R v ARD [2000] NSWCCA 443 at [23], [121]-[124] and [201]; R v PLV [2001] NSWCCA 282 at [21]-[34].

In some cases a jury has acquitted on counts which were not the subject of the original complaint. Although such a discrepancy is capable of affecting credit, it has not been found to require an acquittal on other counts.

In R v JJT (NSWCCA, 3 December 1997, unreported) a factor distinguishing the acquittal and the conviction was that the day after the alleged incidents, the complainant went to the police but lodged a complaint only in terms of the count on which the jury convicted but not in terms of the other count. Gleeson CJ said: “True it is, as the jury were instructed, that the complaint did not amount to corroboration. Nevertheless, the question of consistency was one which the jury were entitled to regard as important and they were entitled to regard the complaint that was made on the day following the events as inconsistent with the evidence in relation to count 3, but as consistent with the evidence in relation to count 2.”

In R v ACK [2000] NSWCCA 180 a similar discrepancy arose. The original complaint did not include reference to the incident on which the jury acquitted. No relevant inconsistency arose ([51]-[53]).

In Asplin v The Queen (WASCA, 26 August 1999, unreported) a number of not guilty verdicts were distinguished on the basis that they had not been included in the original complaint ([26] and [34]).

In Williams v R [2000] TASSC 182 the appellant was charged with six sexual offences committed within one three and a half hour period in the complainant’s home. The jury returned unanimous guilty verdicts on the first count (indecent assault) and the sixth count (vaginal rape). The jury acquitted on the remaining counts (digital penetration, oral rape, and two counts of vaginal rape). The acquittal on the count involving oral rape was a unanimous acquittal, the other three acquittals were by majority. The appellant had admitted the event comprising the alleged indecent assault but denied an absence of consent. He said that none of the other events occurred at all.

Cox CJ, with whom Crawford and Blow JJ agreed, noted that this was a word against word case at [3] and said: “[4] The jury was rightly told to examine each count and not to convict the appellant thereon unless satisfied beyond reasonable doubt of his guilt on that count. The findings they made are consistent with their having been satisfied that the acts comprising counts 1 and 6 occurred without her consent, but that they were not satisfied to the requisite degree that the acts comprising the other four counts had occurred. In the circumstances, it would be remarkable had they been satisfied she had consented to some acts but not to others, two of which were of precisely the same kind, that is, vaginal intercourse, as that on which they had convicted. Indeed, had the only issue been one of consent, then having regard to the close link in time and nature of the alleged conduct, an acquittal on one or more counts would almost certainly have carried an acquittal on the others (see R v Whelan [1973] VR 268, R v Kelly (1985) 38 SASR 561). However, this was not a case where the conduct complained of, other than in respect of count 1, was admitted and the issue therefore confined to consent. The more likely explanation for the differentiation is absence of satisfaction as to the occurrence of all the acts themselves. Disbelief of the complainant’s claims on some of these matters did not logically require them to disbelieve her on all her claims. The doing of the acts comprising count 1 is common ground, consent being the only issue. There was no evidence of recent complaint in respect of digital penetration or the penetration of her mouth, and her complaint that she had been raped did not extend to a complaint of having been raped on several occasions. The verdict is consistent with a jury having been satisfied beyond reasonable doubt that without her consent the complainant had been sexually assaulted and raped at least once. The presence of semen, confirmed by the medical evidence, was capable of corroborating her claim that the appellant had ejaculated inside her on the last occasion she said vaginal penetration had occurred. In my view, there is nothing inconsistent in the jury, satisfied that counts 1 and 6 had occurred in the absence of any evidence other than her own, failing to be satisfied that all the other incidents had occurred, as she had described and giving to the appellant the benefit of such doubt as they had on the matter. There is no affront to logic and commonsense in their verdicts.” [Emphasis added]

One of the factors emphasised in the majority joint judgment in Jones was the proposition that the complainant’s evidence “failed to carry sufficient conviction” in the circumstances where, “her evidence could be set against other reliable evidence” (455). This proposition has been applied, e.g. R v Russell (QCA, 24 March 1999, unreported); R v D [2000] QCA 417 at [12]; R v James [1999] NSWCCA 191 at [64] cf [10]; Eastough v The Queen (WASCA, 12 March 1998, unreported).

However, there are a number of cases where the existence of contrasting or contradictory evidence has been referred to as providing a rational basis for the jury bringing in different verdicts, but the court of criminal appeal concluded that it was open to the jury to convict. This line of authority emphasises the significance of the facts of a particular case.

An express differentiation in this respect was made by this Court in R v AT [2000] NSWCCA 342. In the case of an acquittal on one count on which there was contradictory evidence this was found to be an acceptable basis for differentiation at [79]. (The appeal was successful because of an acquittal on another count which relied solely on her credit ([83]).

Similarly in R v J (No 2) [1998] 3 VR 602 at 628 Winneke P and Charles JA concluded: “… the jury was prepared to give the applicant the benefit of the doubt in respect of particular offences alleged on the presentment where there was acceptable evidence from independent sources pointing to the possibility that the offence had not occurred at the time or in the circumstances particularised in the relevant count. The verdicts, so viewed, do not in our opinion demonstrate inconsistency, nor do they demonstrate that the jury regarded the complainant, on matters of substance , as other than a witness of truth.” [Emphasis added]

The issue in each case may well turn on what, in the circumstances of the case, should be regarded as a “matter of substance”. In a word against word case, cross-examination on surrounding contextual detail is often essential for the defence case. Some surrounding circumstances are of sufficient significance that the Court will conclude that the jury ought to have had a doubt about the complainant’s evidence with respect to other incidents. Not all surrounding circumstances are of such significance.

In Jones, on the count on which the jury acquitted, the appellant’s wife and daughter gave evidence that they were always present at weeknight training sessions and always travelled home with him. On this basis the complainant and the appellant could not have been alone when, as she alleged, he had given her a lift home and the sexual assault allegedly occurred. This was a matter of substance on which the jury did not accept the complainant.

In Jones , the first count of which the appellant had been convicted was alleged, in the end, to have occurred on a Saturday. An assistant coach had said that the appellant drove her home after each such session, but she conceded in cross-examination that it was possible that she had travelled home by train on occasions during the relevant period. This, of course, was precisely the kind of detail affected by the four year delay in the complaint. The jury were not given a Longman warning about the adverse effects of delay, which was particularly pertinent to the evidence of the assistant coach, as emphasised by Brennan CJ, the other member of the majority at 445-446.

The third count, being one of two of which the appellant was convicted, occurred in circumstances where, at least after the delay, no suggestion of third party presence arose and was, accordingly, entirely word against word.

It is important to recognise the significance of the evidence in Jones which contradicted the complainant’s evidence. That evidence was to the effect that no sexual assault of any kind could have happened because the complainant and the appellant were not alone at the relevant time. On the first count the evidence could well have had, but for the delay, the same certainty and would then have been of the same quality as the evidence which led the jury to entertain a doubt about the second count, on which it acquitted. Jones did not involve a conflict over some circumstantial aspect of an alleged assault, cross examination about which could have an effect on whether the complainant’s evidence should be believed, but which did not directly challenge the central allegation of sexual assault.

The case law since Jones indicates that it is not always the case that the jury ought to have had a doubt about the complainant’s evidence with respect to all counts, whenever the relevant distinction between the verdicts of guilty and of acquittal is explicable on the basis that, in the case of the former there was not, and in the case of the latter there was, evidence contradicting or calling into question the version given by the complainant.

It is not desirable to lay down any general principle to determine when the impact of contradictory evidence which appears to have led the jury to entertain a reasonable doubt on one or more counts, ought to have so undermined the credibility of the complainant that the jury should have felt such a doubt about other counts, with respect to which there was no contradictory evidence. This court and other courts have reached different conclusions when faced with such a situation, the result turning on the particular facts of the case.

In R v James [1999] NSWCCA 191 the appellant was found guilty of three counts of indecent assault on a female under the age of sixteen, not guilty on a fourth count, with the jury unable to reach an unanimous verdict on a fifth count. Some twenty-one years had elapsed between the date of the alleged incidents and the trial.

With respect to the count of which the appellant was acquitted, the complainant had given detailed evidence about the nature of a house that was on the property at which the incident was said to have occurred. Notwithstanding the lapse of twenty-one years, the appellant was able to testify that there was no building on the property at that time and to call an independent witness to confirm that proposition. With respect to the count on which the jury was unable to agree there was an odd feature of a most improbable statement which the complainant had attributed to the appellant. Simpson J, with whom Bell J agreed, referred to Jones and said that there was “no relevant distinction between Jones and the present case” [64]. Referring to the matters which led the jury to reach different conclusions her Honour said at [64]: “All that can be drawn from these is that, where there was any positive material to contradict, or create a doubt about, the complainant’s claims, the jury doubted her credibility. As in Jones , if they had a doubt about her credibility on the fourth count (and some, at least obviously had doubts about her credibility on the third as well) then, axiomatically, they must (or ought) to have doubted her credibility generally.”

Simpson J referred at [65] to the passage from MacKenzie at 367 specifically the sentence: “But, if there is a proper way by which the appellate court may reconcile the verdicts, allowing it to conclude that the jury performed their functions as required, that conclusion will generally be accepted.”

Her Honour said, with respect to MacKenzie and Jones : “[66] The key to reconciling the two High Court judgments lies in the opening words of the second sentence in the passage from MacKenzie extracted above. An appellate court will not interfere with a verdict of guilty if there is a proper way of reconciling the different verdicts. In Jones the only reasonable explanation for the differing verdicts was doubt about the complainant’s credibility. That does not provide a proper way of reconciling the verdicts.”

Beazley JA dissented. Her Honour referred to the passage at 366 in MacKenzie with respect to inconsistency, saying “the test is one of logic and reasonableness”. Beazley JA added: “[7] In determining whether verdicts are logical and reasonable, the court is also required to ascertain whether the verdicts can be reconciled or explained in some way. ... . Gaudron, Gummow and Kirby JJ provide a number of examples of how apparently inconsistent verdicts are explainable. Thus, if there is some evidence to support the inconsistent verdict, the appellate court is not to substitute its own view of the facts ... . Another explanation for the verdicts may be that the appellate court considers: ‘… that the jury simply followed the judge’s instruction to consider separately the case presented by the prosecution in respect of each count …’.”

With respect to the count on which the appellant was acquitted her Honour noted the contrary evidence about the existence of a building on the property at the relevant time. Her Honour indicated that she agreed that the jury must have had doubts about this and added: “[10] …A number of possibilities of course present themselves as to the complainant’s version. She may, for example, have made up the incident. Her recollection may have been seriously awry as to the condition of the property. There may have been some such incident which she confused or perhaps even merged different incidents. Whatever be the correct position, there is no doubt that the complainant was not believed on her oath in respect of the incident alleged. This, of course, is a matter which affects her credit. However, the jury had the benefit of seeing and hearing both the appellant and the complainant and clearly believed the complainant on the count in respect of which they returned a guilty verdict. A logical and reasonable explanation of the differences in the verdict is that the jury obeyed, as they were bound to do, the directive that each charge had to be considered on its own and they had to be satisfied of the guilt of the accused beyond a reasonable doubt before being able to come to a guilty verdict.”

With respect to the count on which the jury could not agree, her Honour indicated that a question of credit also arose, although there was no independent evidence contradicting the complainant’s version. Her Honour then said: “[12] … the question arises whether the doubt as to the complainant’s credit in respect of the third and fourth counts so infected her evidence generally that there should have been a reasonable doubt as to the guilt of the appellant on all counts.” Beazley JA referred to Jones and concluded that the case was not directly governed by Jones and in her opinion the verdict was not unsafe and unsatisfactory.

In R v W (1999) 109 A Crim R 51 the appellant was convicted of two sexual offences and acquitted of one sexual offence against the same child during the same period. The Crown case was entirely dependant on the evidence of the complainant (p53). Greg James J, with whom Newman and Sperling JJ agreed, rejected the ground of appeal based on the reasoning in Jones . The count upon which the appellant was acquitted was said by the complainant to have occurred in the presence of the appellant’s then nine year old daughter. That daughter was called to give evidence and denied the happening of the events in count 2 (on which the appellant had been acquitted). She could not recall any circumstances consistent with the allegations comprising count 3. This established a rational basis for the decision of the jury. Greg James J concluded at p58: “True it was that the complainant’s evidence was challenged, not only by that of the appellant but also by the evidence of the appellant’s daughter on count three also … but in my view the evidence of the complainant was not such that her credibility should be considered so tainted that there should be acquittals entered. I do not consider the verdicts were not supported by the evidence or that the evidence was insufficient: Fleming (1998) 103 A Crim R 121. Nor do I consider that the convictions were unreasonable applying the test in James [1999] NSWCCA 191 …”

In R v AG & RJM [2000] NSWCCA 86 one of two appellants was found guilty of three counts, acquitted on one, with the jury unable to reach a verdict on two counts. The other appellant was indicted on four counts and found guilty on all four. One of the counts upon which the jury were unable to agree was said to have occurred on an oval near what the complainant said was a toilet block. The count on which the appellant was acquitted was said to have occurred in that toilet block. Other evidence showed that at the relevant time there was no toilet block.

Simpson J with whom Spigelman CJ and Ireland J agreed said: “[29] The existence of a toilet block was of fundamental importance to proof of the charge in count 8. Its existence was not fundamental to proof of the charge in count 1, but was of some significance as a collateral matter in the assessment of C’s evidence in relation to that count. [30] It is a reasonable inference that the doubts created about the existence of the toilet block were the reason for the doubts in the minds of those jurors who were not satisfied of RM’s guilt on count 1, and the reason for the doubt in the minds of all the jurors in relation to count 8.”

With respect to the other charge on which the jury was unable to reach a verdict, there was other evidence suggesting the presence of other persons at the time and the place where the offence was alleged to have occurred “rendering the description of the events given by [the complainant] unlikely” [35].

Simpson J concluded: “[38] The three charges on which the jury did not convict can be explained by reference to particular objective circumstances I have mentioned - the absence of the toilet block in 1978-1979, and the public nature of the Prospect County Council at the time of the events alleged to constitute count 5. True it is that, in accordance with Jones , it is necessary carefully to scrutinise the jury verdicts, having regard to the possibility that the acquittal and the disagreements are properly to be seen as representing a doubt in the jury’s minds about C’s credibility, a doubt that would necessarily apply to his credibility on other counts as well. I do not think this is such a case.”

Similar issues arose in RM v The Queen (WASCA, 23 February 1999, unreported) . The case involved three counts. With respect to one there was a verdict of acquittal by direction by reason of the absence of evidence. The jury convicted on one count and acquitted on the other.

Murray J, with whom Malcolm CJ and White J agreed, referred to Jones in the following terms: “In the circumstances of that case the High Court found it impossible to discern any distinguishing factual features logically capable of explaining the different outcomes. Indeed, in relevant respects it appeared that the evidence was substantially the same and so their Honours concluded that once the jury found that the complainant’s evidence, which was uncorroborated, lacked sufficient cogency to convict of the offence of which the jury acquitted the appellant, it was impossible to see how they could have arrived at a different conclusion with respect to the counts upon which they convicted.”

In the case before the court, his Honour found that the jury could conclude that “despite some inconsistencies and uncertainties, the child’s evidence was an essentially reliable account in relation to the conduct constituting the element of sexual penetration ...”. With respect to the count of which the appellant was acquitted, the complainant had said the offence was committed in Karratha. She gave evidence that there was a shower block with two cubicles in which the alleged incident occurred. When she was cross-examined there appeared to be inconsistencies between her evidence and police statement about how many showers there were at Karratha.

Significantly, an independent witness gave evidence to the effect that the trip that the appellant made in his employment to the north was unlikely to have included Karratha. Murray J concluded: “In my view it is perfectly feasible to suppose that the jury, being required to deal with a specific incident, found the evidence … to be a significant barrier to their capacity to rely upon the child’s evidence as being not only a truthful recollection, but also an accurate and reliable one, and I can see no reason to suppose that the jury might not reasonably have concluded that they remained in reasonable doubt about that alleged offence, while they accepted the accuracy and reliability of the child’s account of digital penetration. This is not a case, I think, where it must be concluded that the failure of the jury to accept the complainant’s evidence as reliably establishing the offence of indecent dealing imported an incapacity for them to accept her as a witness of truth whose evidence did reliably establish the offence of sexual penetration.”

Malcolm CJ described the evidence of the independent witness as “sufficient to raise a reasonable doubt”.

In R v ARD [2000] NSWCCA 443 with respect to one count, the explanation of the complainant’s presence with the appellant was that she had been asked to feed the birds in the appellant’s de facto wife’s absence. The de facto, who otherwise gave evidence against the appellant, said that she had made no such arrangement about the birds. I concluded at [28]: “In view of this direct contradiction between RD Junior and RD Senior with respect to the circumstances of Count 11, there was in my opinion a relevant basis for differentiation upon which the jury could distinguish Count 11 from other counts. That differentiation was based on the surrounding circumstances of the specific incident. It was not of a kind which damaged the credibility of RD Junior on all counts.” Adams J said at [202]: “I think it likely that the jury accepted that there were a number of sexual assaults perpetrated on RD by the appellant but, where there was apparently credible evidence which showed that a closely connected fact asserted by her to be part of the sequence of events may not have occurred, it was not prepared to convict. That was sufficient to give rise to a reasonable doubt. So far as count 11 was concerned, the contradictory evidence was rather slight, which indicates, perhaps, that the jury’s judgment was a somewhat finely balanced one.”

In R v J (No 2) the Court of Appeal of the Supreme Court of Victoria had before it a case in which sexual offences had allegedly been committed by a father on his daughter over a period of 21 years, when the complainant was aged between eight and 29. There were three counts of indecent assault and ten counts of incest. The accused was convicted on three counts of indecent assault and four counts of incest and acquitted on the other six counts of incest. Winneke P and Charles JA in a joint judgment with which Callaway JA relevantly agreed concluded at 628: “… it seems to us to be clear from the verdicts that the jury accepted the complainant as a witness of truth when she testified that she had been the subject of sexual abuse from the applicant …. However it is also clear from their verdicts that the jury was prepared to give the applicant the benefit of the doubt in respect of particular offences alleged on the presentment where there was acceptable evidence from independent sources pointing to the possibility that the offence had not occurred at the time or in the circumstances particularised in the relevant count . The verdicts, so viewed, do not in our opinion demonstrate inconsistency, nor do they demonstrate that the jury regarded the complainant, on matters of substance, as other than a witness of truth. On the contrary they demonstrate that, in respect of the charges upon which they convicted the applicant, they must have accepted the complainant as a witness of truth and accuracy. It is necessary to bear steadily in mind as Hayne JA and Southwell AJA pointed out in R v Girgines (unreported, Court of Appeal, 26 March 1996) that the jury’s verdicts of acquittal on some counts do not amount to a positive finding by the jury that the events as recounted by the complainant did not occur. As we have already pointed out, the verdicts of not guilty show no more than that the jury was not satisfied to the requisite standard that the acts alleged in those counts occurred or occurred at the time or in the circumstances particularised in them.” [Emphasis added]

In R v R [2000] QCA 27 the appellant was convicted of two counts of sexual offences with a child and found not guilty with respect to six counts. The complainant was ten or eleven at the time of the alleged offences which occurred over a period of almost two years. McMurdo P said: “[26] I am satisfied there was a rational basis open for the jury to reach its verdict of guilty on counts 6 and 7, whilst acquitting on the remaining counts. As to count 1 there was evidence from the appellant’s mother that at all times when the complainant visited her she was present with the appellant and the complainant. The complainant was unsure as to when it occurred and failed to give the evidence opened in respect of count 2. This may well have caused the jury to have a reasonable doubt in respect of her evidence as to count 1. [27] As to counts 3 and 4, the complainant’s brother Damien said in cross-examination that the complainant was picked up after Damien met up with the appellant at the boat in dry dock, not before. This may well have left the jury in doubt about the complainant’s evidence on this count. [28] As to count 5, there was no evidence supporting the complainant’s version of what was a brazen and bold incident in a small boat with two others in close proximity, neither of whom noticed anything untoward. The jury may reasonably have decided that they could not accept the evidence of the complainant beyond reasonable doubt on this count. In not being satisfied beyond reasonable doubt of the complainant’s evidence on these counts they were not necessarily deciding she was not credible.”

The fine line often involved in resolving these issues is illustrated by two contrasting decisions of the Supreme Court of Western Australia: Eastough v The Queen (WASCA, 12 March 1998, unreported) and Wimbridge v The Queen [2000] WASCA 117. The different results is a manifestation of the proposition that each case must turn on its own facts.

In Eastough, the applicant was indicted on nine counts involving sexual offences against his daughter over a period of six to seven years. He was convicted of seven counts and acquitted of two counts, one of sexual penetration and the other of indecent assault. Kennedy J, with whom Steytler J agreed, applied Jones . His Honour said: “Appeals based on inconsistency of verdicts commonly fail by reason of the fact that there is to be found in the evidence some particular feature which explains why the jury could have distinguished between the evidence on the various counts in the indictment and convicted the accused on some counts, while acquitting him on others. There is a relevant feature in this case, but it is that, in relation to the counts on which the appellant was found not guilty, there was present in the house a third person, the appellant’s then wife, who gave evidence which contradicted that of the complainant. In relation to the counts on which the appellant was acquitted on his trial, the complainant was very precise in her evidence. The infant Roslyn was sleeping in her cot in her bedroom. She described the appellant’s coming in and waking her up and whispering her name, and she gave detailed evidence as to the complainant’s other actions and as to the surrounding circumstances. The evidence of the complainant was that these offences occurred before the appellant’s son Timothy was born, at a time when Roslyn and the complainant were sleeping in the middle room in the appellant’s house. The evidence of the appellant’s then wife was that, before her son Timothy was born, Roslyn had always slept in the main bedroom, and, furthermore, there had never been any occasion when she had been awakened by Roslyn’s crying when she found that the appellant was not in his bed. It is therefore clear that the complainant’s evidence was not accepted where another witness was available to give evidence throwing significant light on the surrounding circumstances … . In this case, as in Jones v The Queen , implicit in the appellant’s acquittal on counts 5 and 6 was a non-acceptance of the complainant’s account of the events which were said to give rise to these counts, and her overall credibility was diminished. The only reasonable conclusion is that the jury were not satisfied beyond reasonable doubt of the truth of her evidence concerning these counts. There was nothing to give any cause for supposing that her evidence was more reliable in relation to the other counts than it was in relation to these counts.”

Pidgeon J dissented. His Honour referred to the fact that the test involved an allegation of continuous misconduct from an early period in the life of the complainant by her father over an extended period of time, in contrast with an allegation of misconduct over a shorter period by an unrelated person (as was the case in Jones ). His Honour concluded: “I do not consider on the facts of this case that it would be a proper judgment for this Court to make that the acquittal damaged the credibility of the complainant in respect of her evidence of the acts of impropriety. The complainant was having difficulty in putting them into a time frame and the acquittal is consistent with a doubt of them occurring in the circumstances outlined by her when describing the fifth and sixth counts. She could have been mistaken on the fact that her stepbrother had not been born. This would give rise to a reasonable doubt as to whether the conduct occurred in the circumstances outlined by her.”

In Wimbridge the applicant had been convicted of two offences of sexually penetrating his daughter but was acquitted of a third charge of a similar nature.

In the case of the third count, on which the acquittal was entered, the complainant described the offences occurring on the applicant’s boat shortly after the complainant’s aunt who was visiting from Melbourne had left the boat. The effect of the aunt’s evidence was that the only time that she visited the boat with both the applicant and the complainant was some years before the alleged date of the offence. Pidgeon J, with whom Wallwork and Murray JJ agreed, concluded: “[58] I consider that the convictions on the other two counts can be explained on the basis that the jury did not believe the applicant’s evidence generally by reason of the first two convictions. The jury had been instructed to look at the complainant’s evidence with great care and to treat each matter separately. The jury were then required to assess the position that a witness called by the Crown to support the complainant’s evidence as to the date a vessel was visited and the identity of the vessel said no visit was made by her with the complainant on the date in question. Her evidence was the visit described by both the complainant and the applicant was at an earlier time and prior to the applicant acquiring the vessel Pretty Woman . His Honour referred to the fact that the aunt’s evidence did not support the complainant. It did not do so because the aunt’s evidence showed that the only time that they were on a launch together was at a much earlier time. I consider the aunt’s evidence as to time would be a likely cause for the jury not to convict having regard to the directions they received and it is not an indication that the complainant’s overall credibility was diminished.”

This review of the authorities indicates that a reasonable jury is not obliged to have a doubt about the occurrence of some incidents about which it has only a complainant’s uncorroborated evidence, if they have such a doubt about other incidents whenever, in the latter but not the former case, there is evidence contradictory of, or failing to support, the evidence of the complainant. Any conclusion about this matter depends on the full range of relevant circumstances. Insofar as previous decisions of this Court such as RAT, Barnett, Carbone, Dixon and S, suggest the contrary (and the actual results in those cases depend on their particular facts) they should not be followed. The Evidence on Count 6

In the present case, count 6 alleged an act of sexual intercourse without consent. Counts 1 to 5 occurred over the period 29 January 1981 to 24 July 1981. There were no charges relating to alleged incidents later in 1981, although the complainant gave evidence that such conduct continued to occur. Count 6 was said to have occurred on 4 February 1982. At first the complainant said that this was the first day of the school term. After cross-examination she admitted that it occurred on the second day of the school term being a Thursday. All of the evidence indicated that it was on Thursdays and Fridays that her mother came home late from work and, accordingly, the complainant stayed at the Appellant’s apartment.

Each of counts 1 to 5 allege a single identifiable act of indecency (in four cases) and intercourse (in one case). Count 6 differs in this regard. In the light of the extended definition of sexual intercourse, a number of different possible acts were left to the jury as constituting the offence, including penile penetration, digital penetration and cunnilingus.

The complainant’s evidence on count 6 was as follows: “Q. What happened on that day after the accused came home? A. [TM] pointed to the bedroom door and he was pushing my - he pulled his pants down, pulled his penis out. Q. Is this when you were in the bedroom? A. It was behind the door in the bedroom, yes. Q. Which bedroom? A. His, [TM]’s bedroom. Q. What happened? A. I remember I had like my school uniform on and I saw his hand come down to go under my dress and I didn’t want it to happen so I pulled my dress down tight so he couldn’t do it but he kept trying to pull my dress and I ended up with my dress up around my waist. He pulled my pants down and he pulled his penis out and he put his penis to my vagina and to my arse also. I remember his finger - he was putting his finger into my arse and it hurt. I remember that I cried out, I made a sound. I didn’t scream, I cried out a little bit and he told me to be quiet and he still kept fiddling around with his finger and pushing his penis on to me. Q. What do you mean by fiddling with his finger? A. He was putting his finger in and out of my vagina, in and out and around on my vagina and also he was putting it up into my arse as well. Q. Did anything else happen? A. I went into the lounge room and he pointed to the bedroom again and told me to get under the bed, so I went under the bed. Q. How were you positioned under the bed? A. I went around the side of the bed and my head went under the bed first and my legs were sticking out of the bed. My legs were straight and they were closed and he pulled my legs apart, trying to part my legs. He pulled my pants down and then he licked me on the vagina and he was pointing his tongue into my vagina and putting his finger in, so he would put his tongue first and then his finger, then his tongue and his finger and it was very painful. It hurt me a lot. He was forcing his finger in and it was very rough. Q. Did you do anything to try and stop him? A. I was pushing him on the head telling him to stop. Q. What did he do? A. He kept going for a while and then he stopped. Q. What did you do? A. I got out from under the bed, pulled my pants up myself.”

The complainant had earlier given evidence about another incident that had occurred at the end of February 1981, being an incident about which there was no count in the indictment. She had given evidence that this incident had also occurred under the same bed in the following terms: “Q. When you went into that bedroom, were you told to go somewhere? A. He told me to go under the bed. Q. What sort of bed was this? A. It was just a mattress but there was nothing under the bed. It was very low to the floor, the mattress. Q. And was it a single mattress or a double mattress? A. Double. Q. You said you were told to go under the bed? A. Yes, he kept pointing to get under the bed. Q. And did you do that? A. I said, “No.” I pointed to the other side and he kept pointing to going under the bed and he grabbed my arm and told me to get under the bed. Q. And did you go under the bed? A. I did. Q. And how much of your body was under the bed? A. My legs were out but my body was under the bed. Q. How much space was there between the floor and the bottom of the bed? A. Not much, the mattress - there was probably, maybe an inch and a half, two inches between my face and the bottom of the mattress. But I was on my front and he turned me around and my legs were sticking out. Q. Did something happened whilst you were partially under the bed? A. He pulled my dress up. [TM] pulled my dress up and pulled my pants across and looked at my vagina. Q. Did you place his right little finger which had a long fingernail --? A. Yes. Q. -- after he wet his finger with his own saliva into your vagina? A. Yes, he did. Q. Can you describe what happened? A. First of all he used his left hand to pull my pants across to look at my vagina. Then he wet his finger with his own mouth, inserted his finger into my vagina and was scratching me and I pulled, put my hand down on to his head to push him away but he ignored me and just continued to do it.”

In his summing up, the trial judge left count 6 to the jury in the following way: “The sixth count, that is the 4 February 1982 date. Again, according to the complainant, they were playing. The accused pointed to the bedroom door and pulled his pants down, pulled his penis out. And according to her, ‘he pulled my pants down and pulled his penis out and put his penis to my vagina and to my arse. I remember him putting his finger into my arse too’. Then she said she cried out. ‘I made a scream. I didn’t scream. I cried out a little bit’, and so on. Then; ‘He putting his finger into and out of my vagina and in and out and around my vagina. Also he was putting it up and into my arse as well.’ After that activity - that is what the Crown relies on in support of the charge of sexual intercourse without consent, count number six. The Crown also relies on evidence that after the activity I have just described, he engaged in cunnilingus and licking her in the vaginal area using his tongue and putting his finger in and alternating between his tongue and finger. She said, ‘It was very painful. It hurt me a lot. He was forcing his finger in, very rough.’ So those are the various physical activities the Crown relies on in support of each of the six counts.”

The cross-examination of the complainant included a number of specific matters relating to particular charges and suggestions of inconsistency between the original statement that she made to the police and the evidence she gave in court. The thrust of the cross-examination concentrated on one general theme and one specific theme.

The general theme was the improbability of the events occurring in circumstances of likely discovery. On the complainant’s version of events at all times the appellant’s wife was present in the apartment and cooking in the kitchen. Furthermore, with respect to a number of counts, the alleged incidents occurred during the course of a game of hide and seek in which the appellant’s then four year old son was a participant. On no occasion did he disturb the appellant during the alleged incidents, some of which occurred over several minutes, even though he would have been looking both for the appellant and the complainant in a small flat after counting for ten seconds.

The specific matter related to count 6 and concerned the bed which featured in one aspect of the alleged incident. The defence case asserted that the bed was still owned by the appellant at the date of trial and used in his new residence. This was in the form of an ensemble bed with a base very close to the ground and a mattress on top of it. That bed clearly would not permit any person, even a young child, to lie underneath it. Another focus of the cross-examination was on the improbability of the event occurring as the complainant testified, even if the bed had been of the nature described by the appellant with an inch or two above her face when she lay underneath it.

The appellant’s submissions in this Court focused on the evidence concerning the nature of the bed present in the main bedroom at the appellant’s family’s apartment. Although the appellant’s evidence in this regard was expressed in a number of ways, he, his wife and his son each asserted that the bed which was present in that bedroom in 1981 and 1982 was taken by the family to their new residence and was the same bed that they still had. Photographs of that bed showed a gap of only a few inches between the floor and the bottom of the bed and, accordingly, the events as recounted by the complainant as to part of count 6 could not have occurred. She could not physically have been under the bed in the way she asserted, if the bed was of the character described by the appellant’s witnesses.

Count 6 alleged an act of sexual intercourse but there are a number of different acts capable of constituting the act of intercourse. The events said to have occurred behind the door on this occasion included penile penetration of both the vagina and the anus as well as digital penetration of both the vagina and the anus. Events said to have occurred under the bed on the same occasion included cunnilingus and penile penetration of the vagina.

It does not appear that the Crown was confined by particulars to any specific act of penetration as constituting the act of intercourse for the purposes of count 6. The trial judge’s summing up, to which no objection was taken, left a number of these alleged acts to the jury. Accordingly, it was open to the jury to find the appellant guilty on count 6 on the basis of events that allegedly occurred behind the door whilst setting aside the events which allegedly occurred under the bed. The fact that they found the appellant not guilty on count 6, suggests that they may have been concerned with some aspect of the case which went beyond the conflict of evidence about the bed.

In its submissions the Crown identified a factor distinguishing count 6 from other counts. Count 6 was specific in terms of the date alleging that the offence occurred “on 4 February 1982”. Two of the other counts used the formulation “on or about” particular dates. The other three counts which used a formulation of “on” a particular date did not give rise to any dispute as to the date. As Simpson J notes, nothing in the summing-up suggests that any issue arose during the course of submissions with respect to the evidence on this matter. Nevertheless, the evidence was not entirely satisfactory and could well have given rise to a question in the minds of the jury. The jury received no direction to the effect that the date was not essential.

In the course of her evidence in chief the complainant answered “yes” to the question: “Was 4 February 1982 the first day of the new school term?”. She then gave the evidence as to the events that transpired relevant to count 6. Her evidence, as confirmed by other witnesses, was that it was only on a Thursday or Friday that she spent time at the home of the appellant. In the course of cross-examination she admitted that Wednesday rather than Thursday was the first day of school and confirmed that it was only on the Thursday that she was at the apartment. Her attention was drawn to her statement to the police where she had asserted, as she had asserted in her evidence in chief that it was Thursday 4 February that was the first day of a new school term. She agreed that that was what she had told the police and then was asked the following and answered as follows: “Q: Was what the (sic) told the police accurate? A: No, it was the right information.” It is not clear what this meant.

In the event, after cross-examination, it was clear that she said that the alleged offence occurred on the Thursday, which she now accepted was the second day of the school term.

The difficulty for present purposes is not the minor change to her evidence as to whether or not the incident occurred on the first or second day of the school term. This Court was informed, and it may be that the Court could take judicial notice, that 4 February 1981 was, in fact, a Thursday. However, the jury did not know that. It may have been left in doubt as to whether the date of the offence being a Thursday, was as specified in the indictment, being 4 February.

No-one identified to the jury that the date was not an essential aspect of the charge (see R v KWG [2000] SASC 398). In the circumstances, where the jury did not have any evidence before it as to whether or not 4 February was the first or the second day of the school term, it may have concluded that the Crown had not established that the alleged offence occurred on the date specified in the indictment. This is a possible explanation for the acquittal on this count.

This would be enough to dismiss the appeal on this ground. As a different view may be taken in this regard, it is appropriate to consider other bases for differentiation between the divergent verdicts.

Even if the jury did have some doubt based on the evidence of the appellant, his wife and the son about the identity of the bed, the not guilty verdict does not necessarily indicate that they ought to have had a doubt about the complainant’s evidence of a character which would have required them to have a doubt about her evidence with respect to the other counts.

It is possible for the jury to have decided that a conviction of the appellant on five out of six counts was sufficient to reflect the total criminality of his conduct and that there was no point in deciding the issue as to whether or not the contrary evidence did give rise in their minds to a reasonable doubt with respect to the complainant’s evidence about this count based only on the question of the nature of the bed.

The evidence in relation to the bed included the assertions by the appellant, his wife and the son of the continued identity of the bed in its new location, which established that it was not possible for even a young girl to lie underneath it. There was, however, evidence from a police officer that when she inspected the bed at the new apartment she formed the impression that it was “new”. The police officer was cross-examined to establish that she had not undertaken the kind of inspection that would enable her to make such a judgment. Nevertheless, no other attempt was made to establish in any objective way that the bed was in fact an old one. The jury could well have concluded that, the issue of age having been raised, the police investigations should have determined the age one way or the other. In the absence of any such evidence, the jury may have decided that the appellant was entitled to the benefit of the doubt. That does not mean that the jury’s failure to accept the complainant’s recollection about the bed ought to have led them to doubt her credibility generally.

There is another explanation for the not guilty verdict with respect to count 6 which, in my opinion, is a more likely explanation of the jury’s differentiation amongst the counts. There was a difference between the complainant’s evidence about count 6, which may satisfy the description of a differentiation in the “quality” of her evidence in accordance with the reasoning in Jones , albeit one which raises its own issue of credibility.

The complainant’s evidence with respect to the essential aspects of the acts constituting the offence for the first five counts was as follows: · Count 1 - touching and biting a breast behind the bedroom door. · Count 2 - touching a breast behind the bedroom door, return to the loungeroom and then back to the bedroom door and an act of masturbation. · Count 3 - rubbing of the penis in the toilet, then moving to the adjacent bedroom where an act of fellatio occurred. · Count 4 - rubbing of the vagina while sitting on the balcony, then going to the bedroom where an attempt was made of penile penetration and actual digital penetration occurred. · Count 5 - act of fellatio behind the bedroom door.

Part of the cross-examination of the complainant was directed to establish the inherent improbability of many of the events that were said to have occurred by reason of the likelihood of discovery. On all relevant occasions the appellant’s wife was in the kitchen and on most of the occasions the then four year old son of the appellant was present and on many occasions in the course of playing a game of hide and seek. Finally with respect to the incident alleged on the balcony the evidence suggested that it may have been possible for neighbours to observe the acts said to have occurred there.

Each of counts 1-5 represented a limited number of sexual acts capable of being completed within a comparatively short period of time. Count 6 was, in this respect, different. During the course of the events said to constitute count 6, the complainant alleged: · penile penetration of the vagina. · penile penetration of the anus. · digital penetration of the vagina. · digital penetration of the anus. · cunnilingus alternating with further digital penetration.

It was open to the jury to find that it was inherently unlikely that all of these events had occurred at the same time. The existence of such doubt would not necessarily affect the jury’s assessment of the complainant’s evidence on the other counts. There was evidence of sexual assaults on occasions other than those specified in the six counts before the court The jury may well have formed the view that the complainant had telescoped a number of different events into a single event on that particular day which was the subject of count 6. This could raise a question in the jury’s mind as to the reliability of her evidence with respect to that specific count, being a question which did not necessarily lead them to have a doubt as to her honesty, or otherwise affect their preparedness to accept her evidence with respect to the other counts.

As in the case of a doubt about the identity of the bed, the possibility that she had telescoped a number of different events into one, was capable of affecting her credit on other counts. Whether it did so was quintessentially a jury question. Having carefully read the whole of her evidence, I have concluded that it was open to the jury to accept the balance of her evidence to the criminal standard, while entertaining a doubt, if that is what happened, about her evidence relating to count 6.

The complainant was deaf and gave evidence by signing. The transcript suggests that all her evidence was given with assurance and, given the lapse of time, with an extraordinary range of detail. In this regard nothing differentiated the evidence on count 6. Nonetheless, the lapse of time was such as to affect anyone’s capacity to recall. It was open to the jury to conclude that some aspect of the fragility of memory had affected the complainant’s recollection with respect to count 6, but did so for that count alone. The clarity and conciseness of her evidence with respect to counts 1-5, as it appears to me from the transcript, carries conviction. It obviously did so for the jury who saw her give her evidence. In my opinion, it was open to the jury to accept that evidence, even if it had a doubt about her evidence on count 6 of the character I have discussed.

When it comes to assessing the alleged inconsistency in the overall context, further points of distinction appear between this case and Jones .

One aspect of the evidence in Jones which was emphasised in the joint judgment, as quoted above, was the fact that the delay in that case was unexplained. That is not the case here. The complainant indicated that at one stage the appellant had made a sign to her indicating that he would cut her throat, telling her to be quiet. This occurred after the incidents which comprise counts 1 to 3.

In her evidence, the complainant admitted that on all occasions the complainant’s wife (VM) was in the kitchen and she did not complain to her. The appellant said in re-examination: “Q. Why didn’t you go straight to [VM] and complain to her? A. Because he had threatened. He had gestured to me the cut throat gesture and told me to ‘shush’ putting his index finger over his mouth and also because I was afraid my mother would smack me because - and that’s why I just didn’t say anything to anybody. Q. Before he made that gesture about cutting your throat do you recall the occasions when you say the accused did things to you? A. Yes. Q. Why didn’t you tell [VM]? A. I was afraid I was a young child. Q. Why didn’t you tell your mother about any of these things? A. I didn’t want to. Q. Why? A. My mother would smack me; she smacked me often.”

Although no immediate complaint was made after counts 1-3, the victim was a young girl with a significant disability. Thereafter, unlike Jones , an explanation was proferred for the absence of complaint.

Secondly, as I have indicated above, an important feature of Jones was the fact that there was no Longman warning. In the present case there was such a warning. The emphasis which the judgment in Jones gave to the significant effect of delay upon the ability of the appellant to defend himself, in the context of reaching the ultimate conclusion that it was not open to the jury to convict was, in my opinion, affected by the absence of such a warning. That is not the case here.

Thirdly, in Jones the evidence of the complainant was entirely uncorroborated. In the present case, with respect to count 2 the complainant said that she remembered the acts of forced masturbation had hurt her because the appellant had a long fingernail which was scratching her hands and her arm and she told him that it hurt. Furthermore, in cross-examination about count 4, she said that the act of digital penetration hurt her, because of the length of the appellant’s fingernail.

In his Record of Interview with the police the appellant admitted that he did in fact grow a fingernail long enough to remove staples at work, although in his evidence at the trial he sought to describe his fingernails as “normal” in length. In this respect the jury could accept that with respect to counts 2 and 4 that there was corroboration of the complainant’s evidence.

I would dismiss this ground of appeal. Alleged Defects in Directions

The trial judge told the jury that the delay of almost twenty years that occurred in this case between the events and the trial adversely impacted upon the ability of the defendant to test the evidence against him. He gave, as an example, the issue that had arisen as to the nature of the bed present in the main bedroom at the date of the alleged incident in count 6 and the other uncharged incident that allegedly occurred under the bed. He pointed out that if delay had not occurred an accurate date of manufacture could have been established. He concluded by directing the jury: “If you are of the view it was difficult to test the complainant’s evidence because of the passage of almost twenty years, then you are entitled to take the view that it would be dangerous to convict on that evidence alone unless, scrutinising the evidence with great care considering the circumstances relevant to its evaluation and paying heed to the warning I have given you, you are satisfied of its truth and accuracy.”

This is a warning in accordance with the authority of the High Court in Longman v The Queen (1989) 168 CLR 79 as affirmed in Crampton v The Queen (2000) 75 ALJR 133; [2000] HCA 60. No complaint is made in this case of the warning given in this regard. The warning itself highlights the centrality of the credit of the complainant. Complaint was, however, made as to his Honour’s failure to give directions with respect to other matters relevant to the jury’s assessment of the complainant’s credit.

In his opening remarks to the jury, the trial judge had emphasised that the jury must look at the six charges separately. His Honour indicated in his summing-up to the jury that it was open to them to accept some parts of a witness’s evidence and reject other parts. He also said: “… the Crown is totally reliant on the complainant’s evidence. It is a matter of commonsense as well as law that you would need to scrutinise her evidence with great care before you decided that a verdict of guilty was the appropriate answer and you would certainly need to be satisfied beyond reasonable doubt as to the truth and reliability of the evidence that she gives.”

His Honour also said: “It is up to you how you deal with the evidence and how you find your verdicts, but you may consider that it would be very difficult to find him guilty on one and not guilty on others. You either might accept the complainant as a truthful, accurate, credible witness or you do not. It is difficult to see how.”

With respect to the delay in complaint his Honour gave the jury the statutory direction in the following terms: “The law requires me to give you a direction that an absence of complaint in respect to the commission of an alleged offence by the person on whom the offence is alleged to have been committed, or delay by the complainant in making such complaint, does not necessarily indicate the allegation was false. I am directed to tell you that there may be good reason why the victim in a sexual assault would delay in making …a complaint about the assault. … You remember the evidence the complainant gave about it. Firstly, she said that she was frightened of being smacked by her mum and also she was frightened because of the accused saying ‘shush’ and making the motion across the throat sign. That was the reason she gave.”

Immediately upon retiring the jury sent a note which asked the following: “We would also like to propose a question to [the complainant] as to the reason for coming forward and reporting an incident that occurred several years ago.”

His Honour directed the jury in the following terms: “As to your final inquiry that you would like to propose a question to [the complainant] as to the reason for coming forward and reporting an incident that occurred twenty years ago, I have given you a bit of a direction that we are required to go in in that respect. Again, the difficulty is that [the complainant] is not here and the whole of the evidence in the case is finished. We are, in effect, stuck with that situation. The Crown is not in a position to call her back and certainly would not want to call her and it is not for me to put her back in the witness box after the case is finished insofar as the evidence is concerned. So I am sorry to be so negative, members of the jury, but that is about all I can say.”

In the course of his directions the trial judge indicated to the jury that matters of consistency were of significance when assessing the credibility of the witness. After the jury retired it asked for clarification of the dates on which the complainant had made statements to the police.

The appellant submitted that the jury’s question about the “reason for the complainant coming forward” in substance raised an issue as to whether or not the complainant had a motive to lie. Where such an issue has arisen during the course of the trial by reason of conduct, usually by the prosecution but sometimes by the defence, it will often lead a trial to miscarry (see Palmer v The Queen (1998) 193 CLR 1; R v Jovanovic (1997) 42 NSWLR 520). There are dangers involved in any line of reasoning that suggests a reversal of the onus of proof. It was submitted that the trial judge should have directed the jury in this regard.

The question that the jury posed was not in terms, nor in my opinion in substance, a question which suggested an impermissible line of reasoning on the part of the jury. It was, in my opinion, a natural question which does not necessarily point in one direction or another as far as the impugned reasoning process is concerned.

The jury’s question was not, in my opinion, directed to the proposition that the credibility of the complainant could be enhanced by the absence of a motive for the complainant to lie. The question was probably directed to exploring why the complainant had brought upon herself the inconveniences and, perhaps, the humiliation of a public process with respect to such matters after the passage of so long a period of time. This is a natural question to ask in the context of an assessment of credibility. Of itself, it does not suggest a reversal of the onus of proof.

The form of the question does not indicate in any way that the absence of an answer would redound to the benefit or disadvantage of the appellant. In my opinion this ground of appeal should be rejected.

The appellant made two distinct criticisms of the directions with respect to the jury’s assessment of the credibility of the complainant.

The first criticism was the failure of the trial judge to expressly inform the jury that the delay in the making of the complaint was itself a matter to be taken into account in assessment of the credibility of the complainant. The second criticism was the failure of the trial judge to direct the jury as to the assessment of the complainant’s credibility if the jury were minded to acquit on any count.

As noted above, the trial judge gave the jury a direction as required by s107 of the Criminal Procedure Act 1986 (previously s405B of the Crimes Act 1900). There is authority that such a direction, which tells the jury that there may be reasons for delay in the complaint, should be balanced by the jury being informed that it was entitled to take into account the delay in assessing the complainant’s credibility (see Crofts v The Queen (1996) 186 CLR 427). Indeed authority supports the proposition that such a direction should be given “as a general rule” ( Kilby v The Queen (1973) 129 CLR 460 at 465; R v Davies (1985) 3 NSWLR 276 at 278). Nevertheless in Davies the appeal was dismissed on the basis that in the particular circumstances of the case the issue of the complainant’s credibility had been adequately put before the jury (see 279).

In this case the central significance of the credibility of the complainant must have been obvious to the jury. His Honour informed them that they needed to be satisfied beyond reasonable doubt “as to the truth and reliability of the evidence that she gives”. His other directions reinforced this proposition.

It may be reasonably obvious to most persons that delay in coming forward in the making of a complaint is required to be taken into account in assessing credibility. Nevertheless, in a case such as the present, where the delay is of a lengthy period, in my opinion, a direction of this character should have been given to ensure a fair trial.

In view of the clear direction that the Crown did depend on the complainant’s evidence and the strong Longman warning, I may have been minded to reach the same conclusion as in Davies that the complainant’s credibility had been adequately left to the jury. However, when this omission is combined with the next matter complained of, a different conclusion is appropriate.

The appellant put its submissions with respect to the appropriate direction in the event of a mixture of acquittals and convictions in a number of ways. At first, the appellant sought, but eventually resiled from, a direction in the form propounded by Dunford J in RAT, which I have set out in full above. For the reasons I have given above, such a direction was based on a misunderstanding of Jones .

In part, the direction proposed by Dunford J referred to the consequences on appeal of a finding of a particular character. This is inconsistent with the High Court’s decision in Lucas v The Queen (1970) 120 CLR 171 at 175. Insofar as the direction stated that if the jury finds an accused not guilty on one count, it “ must also find him not guilty on all counts” [emphasis added], it constitutes an impermissible interference with the proper function of the jury. No such direction should be given.

The appellant eventually relied on alternative formulations of a direction requiring the jury to take into account the failure of the complainant’s evidence to satisfy it beyond reasonable doubt in one or more respects on any count, when assessing her evidence on other counts. In this respect the appellant relied on the observations of Fitzgerald JA in R v Robinson (2000) 111 A Crim R 388 at [9]: “The circumstances might be such that the jury should also be instructed that if they hold a reasonable doubt concerning the reliability of a complainant’s evidence on one or more counts, whether by reference to the complainant’s demeanour or for any other reason, they must take that into account in assessing the reliability of his or her evidence in relation to other counts.”

In R v ARD [2000] NSWCCA 441 I referred at [12] and [14], to Fitzgerald JA’s comments in Robinson . I had earlier, in [8]-[9], indicated directions that might be given. It was directions of this character that I had in mind when I said in par [10] that it will often be prudent, and sometimes necessary, to assist the jury in this way. I did not intend to indicate that a direction of the kind proposed by Dunford J as to what a jury must do was appropriate (cf R v KWG [2000] SASC 398 at [8] and [74]).

Fitzgerald JA’s observations in Robinson were made in a context in which his Honour had identified the usual practice of directing a jury to consider each count separately and tell it that it was entitled to accept part of what a witness said and reject other parts. His Honour indicated that directions of this character may not always be sufficient.

As with many aspects of the process of weighing evidence, a judge may be able to assist the jury by drawing its attention to certain matters, even matters which would be regarded by many to be obvious. I express considerable reluctance to add to the number of directions and warnings which trial judges are already required to give. McHugh J noted in KRM v The Queen (2000) 75 ALJR 550; [2000] HCA 11 at [37]: “The more directions and warnings juries are given the more likely it is that they will forget or misinterpret some directions or warnings.”

Nevertheless, the case law on inconsistent verdicts, particularly in the context of sexual assault cases but not limited to that context, indicates that there is a recurring difficulty in this respect. It may appear to be obvious that a reasonable doubt about one aspect of a complainant’s evidence ought to be taken into account when assessing that witness’s evidence on other matters. However, there have been a significant number of cases in which courts of criminal appeal have acted on the basis that the jury may have failed to do so.

In the light of the number of cases it is desirable that the traditional direction as to treating each count separately is supplemented in a word against word case. Some reference ought to be made to the effect upon the assessment of the credibility of a complainant if the jury finds itself unable to accept the complainant’s evidence with respect to any count.

Some form of direction assisting the jury in this respect should be given, to employ the terminology found in Kilby and Davies “as a general rule”. Its absence is not necessarily fatal (as it was not in Davies itself). Furthermore, as the joint judgment in Crofts affirmed, the “general rule” does not apply “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” (at 451).

It is not necessary to specify any precise words for such a direction. That will depend on the circumstances of the case. It will often be appropriate to direct a jury that where they entertain a reasonable doubt concerning the truthfulness or reliability of a complainant’s evidence in relation to one or more counts, that must be taken into account in assessing the truthfulness or reliability of the complainant’s evidence generally.

On other occasions it may be appropriate for a judge to indicate to the jury, whilst making it clear that it remains a matter for the jury, that it might think that there was nothing to distinguish the evidence of the complainant on one count from his or her evidence on another count.

Or it may be appropriate to indicate that, if the jury has a reasonable doubt about the complainant’s credibility in relation to one count, it might believe it difficult to see how the evidence of the complainant could be accepted in relation to other counts.

The precise terminology must remain a matter for the trial judge in all the particular circumstances of the specific case. The crucial matter is to indicate to the jury that any doubt they may form with respect to one aspect of the complainant’s evidence, ought be considered by them when assessing the overall credibility of the complainant and, therefore, when deciding whether or not there was a reasonable doubt about the complainant’s evidence with respect to other counts.

In the present case the trial judge attempted to give a form of direction in this regard. The full quotation indicates that the grammar of the passage was either not precise or, given other apparent defects of the transcription, was not accurately recorded.

The appellant focused on the fact that the passage inverted the proper reasoning process, when his Honour is recorded to have said: “You may consider that it would be very difficult to find him guilty on one and not guilty on others.” The proposition discussed above would require a formulation of: “You may consider that it would be very difficult to find him not guilty on one and guilty on others.”

If the appeal was concerned only with this matter, I would not have concluded that this particular passage was of sufficient significance to allow the appeal on the basis that the trial judge invited the jury to engage in tendency reasoning, without warning it against the dangers of such reasoning. The inversion appears to have been a slip, which passed without comment by counsel at the trial. The very next sentence was “[y]ou either might accept the complainant as a truthful, accurate, credible witness or you do not”. The context prevented an impermissible course of reasoning. Plainly, by acquitting on one count, the jury did not adopt such an approach.

Nevertheless, the jury did not have its attention expressly directed to the proposition that a reasonable doubt with respect to the complainant’s evidence on any count, ought to be taken into account on the complainant’s credibility generally. This would have been pertinent whether or not the doubt in the mind of the jury arose by reason of the evidence about the bed, or the extended nature of the incidents that were alleged to have occurred in the space of a short period or for some other reason.

Again, if the only inadequacy in the summing-up had been this omission, I would not have allowed the appeal on this basis. This Court has no basis for correcting what may have been a transcription error rendering the trial judge’s comments unclear. Nevertheless, he did direct the jury’s attention generally to the interconnection of credibility issues amongst the various charges. In the overall context, particularly the strong Longman warning, this may have been sufficient. However, there are two matters which relate to the critical issue in the case, namely the complainant’s credibility. The cumulative effect of these matters casts a doubt in my mind on the fairness of the trial.

In my opinion, in the circumstances of this trial the jury required greater assistance in two respects than the trial judge gave in the relevant parts of the summing-up, which I have set out above. The combined impact of the absence of directions on the effect on credibility of delay in complainant and of an acquittal on one count, is such that a miscarriage of justice has occurred.

As the High Court said in Crofts supra at 451: “The overriding duty of the trial judge remains to ensure that the accused secures a fair trial.” In my opinion, the appellant did not receive a fair trial. New Trial?

The appellant was sentenced to a period of two and a half years imprisonment with a non parole period of one year for the fifth count. For counts 1-4 the trial judge imposed concurrent fixed term sentences of one year. He will be eligible for release on parole on 12 October 2001. Accordingly, the appellant has served a major part of his minimum term.

I have given careful consideration to whether, in these circumstances, the Court should direct verdicts of acquittal on all counts. The public vindication of the claims of a victim of offences of this character is significant. The appellant was convicted on five counts of sexually interfering with an eight to nine year old child. Whether or not, in all the circumstances, he should be put to trial again is appropriately determined by the executive branch of government. Accordingly, I propose that the appeal be allowed, the convictions quashed and a new trial ordered.

WOOD CJ at CL: I adopt the statement of facts and issues for appeal identified in the judgments of Spigelman CJ and Simpson J. By reference to that material, I shall confine myself to those grounds of appeal in respect of which I wish to add my own observations. Differentiation in verdicts

With the benefit of the extensive analysis by Spigelman CJ of the many decisions, since Jones v The Queen (1997) 191 CLR 439, which have addressed the question whether a jury verdict is “unreasonable or cannot be supported,” within the meaning of S6 of the Criminal Appeal Act 1912 (NSW) or its equivalent in other jurisdictions, I consider it necessary to restate my own position in respect of that provision. That involves some qualification of the passage in my judgment in R v Carbone (2000) NSWCCA 387, which is extracted in that analysis.

That decision, and other decisions to similar effect (e.g. R v RAT (2000) 111 A Crim R 360; R v Barnett (2000) NSWCCA 283; R v Dixon (2001) NSWCCA 39, R v Mills (2001) NSWCCA 48, and R v S (2001) NSWCCA 204), had dealt with the situation where a jury had returned a verdict of acquittal on one or more counts preferred in an indictment charging multiple counts. They were cases where the only direct evidence of the offences was that of the complainant, and where the jury had convicted the accused upon the remaining counts. They appear to have reflected a presumption that, in such a case, absent the positive identification of some fact or surrounding circumstance providing a rational basis for accepting the complainant’s evidence on the counts of which the accused was found guilty, the other verdicts should be regarded by a Court of Criminal Appeal as unreasonable.

The rationale for such conclusion was that the rejection, by the jury, of the complainant’s evidence, in relation to the counts or counts upon which verdicts of not guilty were returned, involved a fatal diminution in that witness’s overall credibility. These decisions, accordingly, required the Crown, if it were to maintain the convictions, to point to circumstances that could positively restore the complainant’s credibility on the remaining counts, or to identify some rational basis for the jury having distinguished between the counts on which it returned verdicts of guilty and not guilty respectively.

Upon reflection, and with the benefit of a comprehensive review of the decisions to date, I have reached the conclusion that this line of decision has read too much into Jones , and has advanced a principle that is altogether too rigid.

A convenient starting point is the decisions of the High Court in MacKenzie v The Queen (1996) 190 CLR 348 where Gaudron, Gummow and Kirby JJ emphasised, at 366: “Where, as is ordinarily the case, the inconsistency arises in the jury verdicts upon different counts of the originating process in a criminal trial, the test is one of logic and reasonableness”.

As their Honours continued, at 367: “… the respect for the function which the law assigns to juries (and the general satisfaction with their performance) have led courts to express repeatedly, in the context both of criminal and civil trials, reluctance to accept a submission that verdicts are inconsistent in the relevant sense. Thus, if there is a proper way by which the appellate courts may reconcile the verdicts, allowing it to conclude that the jury performed their functions as required, that conclusion will generally be accepted. If there is some evidence to support the verdict said to be inconsistent, it is not the role of the appellate court, upon this ground, to substitute its opinion of the facts for one which was open to the jury. In a criminal appeal, the view may be taken that the jury simply followed the judge’s instruction to consider separately the case presented by the prosecution in respect of each count and to apply to each count the requirement that all of the ingredients must be proved beyond reasonable doubt.”

The respect which is properly to be afforded to the constitutional role of the jury, and the advantage which it has in seeing and hearing the evidence unfold was recognised in M v The Queen (1984) 181 CLR 487 at 493, in a passage cited with approval in Jones at 451: “… the question which the court must ask itself is whether it thinks that upon the whole of the evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty. But in answering that question the court must not disregard or discount either the consideration that the jury is the body entrusted with the primary responsibility of determining guilt or innocence, or the consideration that the jury has had the benefit of having seen and heard the witnesses. On the contrary, the court must pay full regard to these considerations”. (emphasis added)

It is with this fundamental respect for the role entrusted by the criminal justice system to the jury, to find the facts according to the evidence, that this court should apply the test enunciated in M at 494 to 495, and in Jones , when it is required to determine whether guilty verdicts are unreasonable and incapable of being supported. That test necessarily requires close reference to the evidence, since it was said in M at 494 to 495: “… where the evidence lacks credibility for reasons which are not explained by the manner in which it was given, a reasonable doubt experienced by the court is a doubt which a reasonable jury ought to have experienced. If the evidence, upon the record itself, contains discrepancies, displayed inadequacies, is tainted or otherwise lacks probative force in such a way as to lead the court of criminal appeal to conclude that, even making full allowance for the advantages enjoyed by the jury, there is a significant possibility that an innocent person has been convicted, then the court is bound to act and to set aside a verdict based upon that evidence. In doing so, the court is not substituting trial by a court of appeal for trial by jury, for the ultimate question must always be whether the court thinks that upon the whole of the evidence it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty.”

In applying the decision in Jones it is not to be overlooked, as was made clear in the judgment at 455, that it was the combination of the credibility factor arising from the acquittal of the appellant on one count, “with the uncorroborated nature of the complaints,” and the effect of the lengthy and unexplained delay in the making of the complaints, that led to the conclusion that the convictions were “unsafe and unsatisfactory” (as the ground was conventionally described before the decision in Fleming v The Queen (1999) 197 CLR 250 and R v Giam (1999) NSWCCA 53).

These two additional factors, their Honours confirmed, made it “necessary” or “essential” for the jury to scrutinise the complainant’s evidence in the case with “considerable” or “great care”. Additionally, there was the circumstance, to which considerable importance had been attached in Longman v The Queen (1987) 168 CLR 79, and Crampton v The Queen (2000) 75 ALJR 133, concerning the considerable difficulty which the defence may face in dealing with allegations made of events occurring many years in the past.

So understood, I do not regard Jones as an authority for the presumption previously mentioned. It does not deny to a jury the right to accept the evidence of a complainant in relation to one event, and to maintain a reasonable doubt about another event described by that witness. It leaves unimpaired the traditional instruction given to juries that they are not obliged to wholly accept or to wholly reject what a witness says, but are free to accept part and to reject part of his or her account - an instruction accepted as perfectly appropriate, if not one that should be given, in KBT v The Queen (1997) 191 CLR 417 at 424, and in KRM v The Queen (2001) 75 ALJR 550, where McHugh J said: “[36] It has become the standard practice in cases where there are multiple counts, however, for the judge to direct the jury that they must consider each count separately and to consider it only by reference to the evidence that applies to it. … The universal giving of a separate consideration warning and the omission of a universal propensity warning indicates that the giving of a separate consideration warning is ordinarily sufficient to avoid miscarriages of justice in cases such as the present. This indication is confirmed by the many cases where juries acquit accused persons of some charges and convict them of others where the presentment contains multiple counts involving the same or similar offences. Indeed, so freely do juries acquit of some charges and convict of others on presentments with multiple counts that appellate courts often hear arguments that there is such an inconsistency in the verdicts that the convictions are unsafe and must be set aside cf Jones v The Queen (1997) 191 CLR 439”.

That his Honour assessed the outcome, which is at the centre of this aspect of the appeal, as a frequent occurrence, an assessment with which I would agree, does of itself tend to indicate that something more than illogical reasoning lies behind the return of different verdicts.

The kind of direction mentioned, and the reasoning behind it, are matters of common sense which tend, if anything, in favour of an accused so far as they discourage tendency reasoning and invite specific focus on the evidence in relation to each count. Moreover, such a direction allows the jury to bring to their task matters such as the conviction, or apparent persuasiveness, with which complainants describe events of which they have a clear recollection, and the lesser certainty which they may display in relation to other events, of which their recollection may be less clear or where there may be a reasonable possibility of them being mistaken, for example as to dates or places.

This to my mind, reflects the practical, commonsense approach to fact finding which is expected of jurors, as well as the necessity, in some cases, for the prosecution to fit an alleged event within the territorial reach of the relevant statutory provision, or within a specific time frame (as may occur where there have been amendments to the Section defining the offence).

In each of these respects, legitimate questions may arise concerning the evidence of a complainant, or a central prosecution witness, which go to the reliability of their recollection or their full awareness of the facts, without in any way reflecting upon their veracity or honesty . Allowance for uncertainty, or imprecision, or lack of persuasiveness in relation to some acts alleged, particularly where the Crown overloads an indictment, does no more than reflect the important difference between reliability and credibility or honesty, a distinction also commonly drawn to the attention of juries.

In my view the decisions which have presumed the demonstration of a fatal blow to the credibility of the complainant, or central witness, where there are different verdicts in a multi-count indictment, have overlooked this distinction and drawn too much from Jones . More helpful is the conclusion expressed in MacKenzie at 368: “It is only where the inconsistency rises to the point that the appellate court considers that intervention is necessarily required to prevent a possible injustice that the relevant conviction will be set aside. It is impossible to state hard and fast rules. ‘It all depends upon the facts of the case’”.

To similar effect were the observations of McHugh J in Osland v The Queen (1998) 199 CLR 316: “[120] When there is no legal or technical inconsistency in the verdicts, a conviction can only be set aside where the inconsistency in the verdicts demonstrates that no reasonable jury who had applied their minds to the evidence could have arrived at the two different verdicts. In such a case the Court of Criminal Appeal sets aside the conviction because it is unsafe or unsatisfactory. [121] In determining whether the inconsistency points to an unsatisfactory conviction, the appellate court must consider the evidence, the issues, and the directions which the jury were given … An examination of the directions, issues and evidence may confirm that apparently inconsistent verdicts are in fact inconsistent and demonstrate that the conviction is unsafe. Just as frequently, however, examination of the issues, evidence and directions may show that apparently inconsistent verdicts are not inconsistent and that there has been no failure in the reasoning process of the jury”.

These passages, to my mind, underline the importance of not assuming too readily that a difference between the verdicts returned demonstrates inevitably a want of credibility in the complainant or central witness. I would add that it may well be that the use of the term “inconsistent verdicts,” to describe the situation where an accused has been convicted on some counts and acquitted on others, may itself be a misnomer, and part of the reason why too much has been read into Jones . In truth, a difference in verdicts may not involve any inconsistency at all, when the case is considered as a whole.

Indeed, this is implicit in the judgment of Hayne J, with whom McHugh and Gaudron JJ agreed in Farrell v The Queen (1998) 194 CLR 286, a case where the accused was convicted of the several offences in respect of which he was charged as the principal offender, and acquitted of the one count in respect of which he was charged as an aider and abetter. There being an absence of evidence of him offering encouragement to the person charged as the principal offender for that count, it was held that the different verdicts were “entirely consistent with the jury not reaching the requisite degree of satisfaction” concerning it.

In the context of a system of justice that requires proof of criminality beyond reasonable doubt, the occurrence of different verdicts is entirely to be expected and is understandable. As is often said of such a system, acquittal does not amount to a positive finding that the act alleged did not occur, or that the evidence of the complainant or central witness concerning it was rejected as a lie, or as lacking in credibility: R v Girgines VICCA 26 March 1996.

What it can involve is proof falling short of satisfaction beyond reasonable doubt of one or more of the elements of the offence charged. In the context of a case involving stringent warnings of the kind required by Longman v The Queen (1989) 168 CLR 79 and by R v Murray (1987) 11 NSWLR 12, or those customarily given where the prosecution case depends essentially upon an accomplice or prison informer, it is often well understandable that a jury would be hesitant to convict without necessarily finding that the complainant or central witness was altogether lacking in credibility.

The basis for the presumption which seems to have been drawn from Jones concerning the impact upon the credibility of the complainant where different verdicts have been returned, lies in three passages in the joint judgment of Gaudron, McHugh and Gummow JJ. The first appears at page 453. “The jury’s finding of not guilty on the second count damaged the credibility of the complainant with respect to all counts in the indictment. Implicit in the appellant’s acquittal on the second count was a rejection of the complainant’s account of the events which were said to give rise to that count.” Their Honour’s later said (also at 453): “It is difficult then to see how it was open to the jury to be convinced beyond reasonable doubt of the guilt of the appellant with respect to the first and third counts. There is nothing in the complainant’s evidence or the surrounding circumstances which gives any ground for supposing that her evidence was more reliable in relation to those counts than it was in relation to the second count.” In the final passage, at 455, their Honours concluded: “In our opinion, the proper application of the test formulated by the majority of this court in M required the convictions of the appellant to be set aside on the basis that they were unsafe and unsatisfactory. Given the jury’s findings on the second count, it was not open to them, on the whole of the evidence, to be satisfied beyond reasonable doubt of the guilt of the appellant on the first and third counts. Once the jury found that the evidence of the complainant with respect to the second count lacked sufficient cogency to convict, the Crown cases on the first and third counts wore a different complexion. For it meant that, when her evidence could be set against other reliable evidence, it failed to carry sufficient conviction to reach the criminal standard of proof. As we have already said, nothing in the complainant’s evidence gave any ground for thinking that the quality of her evidence was higher in respect of the first and third count than it was in respect to the second count.”

Properly understood, I do not understand their Honours to have been enunciating any general principle for the guidance of appellate courts when dealing with a similar ground of appeal. Rather, the practical significance of the different outcomes was considered, in the context of the particular facts of the case (“the surrounding circumstances”) together with the other problems affecting the prosecution evidence, to have led to convictions that could not be supported.

Although Kirby J was in the minority, in my respectful opinion, his Honour was correct when he rejected the existence of any general proposition, or presumption, of the kind which seems to have led to the line of decisions which are under question in this appeal. At 470, his Honour observed in this respect: “Any other hypothesis would require a conclusion that, in all cases of this kind, complainants must be wholly believed or wholly disbelieved. That would put at naught the need for separate verdicts on the separate counts of the indictment and the principle of particularity in the proof of separate offences charged upon such separate counts. That principle necessarily posits the possibility (at least in the way this trial was conducted) that different verdicts may be returned on different counts as occurred here. It would also overlook the practical way in which juries often approach different charges expressed in different counts of an indictment. (83)”

I am reinforced in this view by the reasons of Winneke P in The Queen v KET (1998) VSCA 73 where his Honour observed: “[28]… Jones’s case, in my view, is a case which has to be looked at in the light of its own particular facts, as indeed all cases where this issue raised must be. In Jones’s case three discrete acts of sexual penetration were alleged and the jury convicted the accused on two and acquitted on one, in circumstances where the acquittal, as the judgments made clear, must necessarily on the facts of that case have reflected upon the credibility of the complainant’s evidence on the other two counts. Furthermore, it would seem from the judgments in Jones’ case that special emphasis was laid on the prejudice flowing to the accused from the delay in complaint because it was apparent from the special facts of the case that there was a real risk that the delay had deprived the accused of what the majority judges had called ‘ a cast iron alibi’. [29] However, in my view, it would be wrong to draw from the decision of Jones ’ case the general proposition that, in cases where multiple sexual offences are alleged involving the one complainant, the jury’s acquittal on some of those counts should compel an appellate court to conclude that the jury must necessarily have regarded the complainant generally as an untruthful witness or that her credibility was impacted upon in respect of the counts upon which they had convicted.”

This conclusion also permits proper respect be given to the right of a jury to return what has perhaps been unfortunately referred to as a “merciful verdict”, when faced with an indictment containing a large number of counts. The expression involves something of a misnomer, since a conviction, upon any count in a criminal trial, cannot properly be regarded as “merciful”. However, it is understandable, and in accordance with experience, that a jury faced with a number of counts involving similar conduct, can be permitted without disobedience to their oath, to take a “merciful view” of the facts upon one count.

Similarly, they may be well justified in coming to the conclusion that enough is enough, and to return verdicts of guilty in relation to the strongest of what may be regarded as representative charges selected by the Crown, out of the much broader and longer history of sexual abuse, that is often seen in cases of sexual assault within a family or within an educational, sporting, recreational or religious environment. It cannot be overlooked that a jury drawn from the community will have an awareness of this, and of the regrettable reality of paedophiliac behaviour.

Support for such an approach has been recognised in the oft cited judgment of King CJ in R v Kirkman (1987) 44 SASR 591 at 593: “Sometimes juries apply in favour of an accused what might be described as their innate sense of fairness and justice in place of the strict principles of law. Sometimes it appears to a jury that although a number of counts have been alleged against an accused person, and have been technically proved, justice is sufficiently met by convicting him of less than the full number. This may not be logically justifiable in the eyes of a judge, but I think it would be idle to close our eyes to the fact that it is part and parcel of the system of administration of justice by juries. Appellate courts therefore should not be too ready to jump to the conclusion that because a verdict of guilty cannot be reconciled as a matter of strict logic with a verdict of not guilty with respect to another count, the jury acted unreasonably in arriving at the verdict of guilty.”

That passage was cited with approval in the joint judgment in MacKenzie at 367-368 and the freedom or entitlement of a jury to return a merciful verdict was noted in Crisologo NSWCCA 12 December 1997. An example, in which this Court assumed that the jury had taken the practical step of regarding a conviction on portion of the indictment, as sufficiently reflective of the culpability of the accused, can be seen in Kiskarpati NSWCCA 4 November 1998. Other examples confirming that the option of returning a “merciful verdict” is open, and is often exercised, can be seen in R v Hunt (1968) 2 QB 433 and R v Johnstone (2000) QCA 234.

The view which I favour also preserves for the jury the advantage which it has over a Court of Appeal in having seen and heard the witnesses, including the subtle conclusions which can be drawn from hesitation and other marks of indecision which may flag that, for some counts, the evidence of the complainant was less certain or reliable than for others. As I have already observed, there is an important difference between reliability and credibility which has been recognised, for example, in decisions such as : R v Robinson (2000) 111 A Crim R 388 and ARD (2000) NSWCCA 443.

In the light of the foregoing, I move forward to the conclusion which Dunford J, reached in RAT , concerning the principle which his Honour considered should be derived from Jones . “[46] If I am correct in my understanding of the effect of Jones then I believe consideration needs to be given to the directions regularly given to juries in such trials. Juries are constantly told that they should consider each count in the indictment separately and also that they may accept part of a witness’ evidence and not accept other parts of the evidence of that same witness. But Jones appears to establish a qualification to these directions to the effect that in sexual cases, where the only direct evidence of the commission of the offences is that of the complainant, if they are for any reason not satisfied beyond reasonable doubt that the complainant is telling the truth in relation to one count, it is not open to them to be satisfied to the criminal standard that she (or he) is telling the truth in relation to any other count; and therefore in such cases (that is, where there is no independent evidence of the commission of the offences) if they find an accused not guilty on one count, they must also find him not guilty on all counts. They should also be warned specifically that if they return different verdicts where there are no distinguishing features in the evidence such verdicts are liable to be regarded as a compromise and the guilty verdicts set aside.” ( emphasis added)

For the reasons already stated, I consider this approach to be incorrect so far as it would elevate the conclusion in Jones either to a principle of law, or to a rebuttable presumption. Rather, I am of the view that the fact of differing verdicts is but one aspect of the review which an appellate court must conduct, and that it should take its place alongside all the other circumstances of the case.

There may well be cases where the fact of acquittal on one or more counts will support an argument that the jury looked with real disfavour upon the credibility of the complainant or central witness. This may arise, for example, where: a) verdicts of not guilty were returned on a preponderance of the counts in an indictment; b) there was positive evidence in the defence case to establish that the version of events offered by the complainant or central witness was fabricated in relation to the counts upon which the accused was acquitted, c) where the evidence offered by the prosecution, in relation to the counts on which the accused was acquitted, appears to be fanciful or inherently improbable (R v W (1999) 109 A Crim R 51 provides a possible example); or d) where the complainant or central witness has given versions, in relation to the count or counts upon which the accused was acquitted, which were significantly inconsistent.

In many cases, conversely, it will be possible to identify a possible basis for a differentiation between verdicts, for example where: a) the complainant or the central witness has conceded the possibility of faulty recollection, in relation to the details of the event the subject of the charge on which the accused was acquitted; b) there was a lack of particularity as to time or place in relation to such charge, and where fixing the event at a particular time or place is either of the essence or of central importance ( R v AT (2000) NSWCCA 342 provides an example of a case where, (but for other circumstances concerning another count), the existence of uncertainty as to time would have provided an acceptable explanation; and see R v J (No.2) (1998) 3 VR 602); c) the accused has called positive evidence in relation to the charge or charges on which he was acquitted but not in relation to the others; d) there was corroboration in relation to the charge on which the accused was convicted, but not in relation to the others; e) it is fair to assume that, in relation to a count or counts in respect of which the accused was acquitted, there was room to suppose that the complainant or central witness resorted to a degree of exaggeration in order to reinforce his or her account; f) the acquittals were returned in relation to less serious events, particularly when they were not mentioned in the original complaint ( R v JJT NSW CCA 3 December 1997, provides an example of such a case) g) the history of wrongdoing is lengthy and involves very many incidents over a period of time, or conversely involves a large number of events within a very short time frame such that a faulty recollection as to one or other of them, or some telescoping of them, may be entirely understandable (see R v J (1994) 75 A Crim R 523 at 539/540); or where h) the jury were entitled to conclude, in relation to the event in respect of which the accused was acquitted, that the complainant misunderstood or misinterpreted what it was that had occurred ( R v ACK (2000) NSW CCA 180 provides an example of such a case).

By this selection of examples, I am not to be taken as suggesting that there must inevitably be a search for an explanation for a difference in verdicts. Rather, my purpose is to underline the need for careful consideration to be given to the process which jury deliberations may properly take, as well as the need for setting that in the full context of the circumstances of the trial, including the terms, and the firmness with which Longman , Murray and other directions are given, and the ‘staleness’ or otherwise of the complaints. What is important in this regard is to give full attention to those matters that go to an assessment of credibility, and to recognise that not all of them operate to the same degree.

I am not, accordingly, persuaded that it is necessarily helpful, or indeed essential, to narrow the search for an actual or possible logical basis for distinction. There will be cases where, once the entirety of the case is reviewed, the appellate Court will maintain a comfortable satisfaction as to the verdicts of guilty in relation to the relevant counts, upon the basis of the evidence separately considered concerning them.

Ultimately, the question remains one of fact and degree as to whether the difference in verdicts is such that, as a matter of logic and reasonableness, bringing to account all of the factors which I have mentioned, including the practical approach which juries are entitled to bring to their task, the conviction should be regarded as unreasonable or incapable of being supported upon the evidence.

In the present case I am of the view, having regard to the various considerations identified by Spigelman CJ in his judgment that there is an acceptable explanation for the differentiation between the divergent verdicts returned in this case. It is not one where there was such an integral connection between the counts, or where there were other circumstances present, such as to compel the conclusion that the complainant’s overall credibility was so diminished that the guilty verdicts should, for that reason, be overturned.

I would dismiss this ground of appeal. Credibility directions

The first matter that arises, in this respect, concerns the failure of his Honour to give a direction in accordance with the decision in Crofts v The Queen (1996) CLR 427, by way of a counterbalance to that given in accordance with S 107 of the Criminal Procedure Act 1986.

It was entirely appropriate that his Honour inform the jury, in accordance with this Section, that the absence of a complaint or a delay in making a complaint, do not necessarily indicate that the allegation was false, and that there may be good reasons why a complainant may delay making a complaint, or remain silent about it.

However, it has become customary since Crofts for that direction to be accompanied by a reminder that the inconsistency of conduct, which the High Court assumes to be implicit from the absence of, or a delay in, the making of a complaint by a person who had been sexually assaulted, can adversely affect the complainant’s credibility. Current authority favours the view that such a direction should be given as a general rule: Kilby v the Queen (1973) 129 CLR 460 at 465; R v Davies (1985) 3 NSWLR 276 at 278, R v Omarjee (1995) 79 A Crim R 355 and Miletic SC VIC CA 9 August 1996.

So far as these decisions assume that, as a matter of ordinary human experience, victims of sexual assault will complain promptly of the assault, then I express my reservation as to the correctness of such a proposition, both generally and specifically in relation to children. While in relation to other forms of criminality it might be expected that a prompt complaint will be made by a victim, I am by no means persuaded that the same expectation exists in relation to sexual assault. Both trial experience and an understanding of the dynamics of sexual invasion, an event that involves multiple and complex emotions of shame, guilt, embarrassment, conflicting loyalties, inexperience, fear and threat lead me to suspect that this assumption is flawed. I also express my reservation as to whether judicial insight into such a matter has any advantage over the experience of lay persons, drawn from the community, sufficient to justify a mandatory direction of the kind sought.

Nevertheless, I respect as authoritative the current line of decision both in the High Court, and at appellate level in this State and elsewhere, in respect of this matter. Their effect is to interpret the legislation in a way which would not sterilise complainants from critical comment concerning the impact of delay or silence, so far as their credibility is concerned. In substance, they call for a balancing direction to be given where the particular facts of the case justify that course.

The reservations I have, as to the assumption which underlie the balancing direction, do however underline the importance of the qualifications suggested in Crofts : “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 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. So long as the purpose of the legislation, to rid the law of such stereotypes, is kept in mind, and the terms in which the legislation is expressed are followed, judges striving to assist juries in their consideration of the facts are unlikely to fall into the kind of error that occurred in this case.”

In this regard, it is not to be overlooked that, in Crofts, there was a serious misdirection so far as the trial Judge had assumed that a Kilby direction was no longer permissible, and so far as the jury was directed “as a matter of law” that the lack of complaint about the offences immediately after they happened, would not permit of an inference that the complainant was lying about them. That clearly went too far, and denied to the jury an opportunity of bringing their common sense and experience of life to the case, so far as it concerned this aspect of their evaluation of the evidence.

Understood in this way, and approaching a summing up by reference to whether, in its full context, and in the particular circumstances of the trial, it would be unbalanced for there to be no reference to silence or delay in relation to credibility, my concerns are to a degree alleviated.

The present is a case where a strong Longman warning was given, along with a clear warning concerning consistency or inconsistency of account. It is also one where it may be assumed, both from the terms in which the statutory direction was given, (in particular so far as it was expressed in terms that this “does not necessarily indicate”, and “there may be good reasons”, and from the jury questions concerning the reason for the late complaint, and also concerning the dates of the complainant’s statements, that they were well alive to the possibility that inaction, particularly for the lengthy period involved, could reflect upon the complainant’s credibility.

Similarly to Spigelman CJ, I would not be minded to allow the appeal on this ground were it to stand alone. However, when taken in conjunction with the second matter to which objection has been taken, I have, with some reservations, come to the conclusion that the jury were insufficiently assisted in relation to the complainant’s credibility.

This second aspect of the appellant’s submission concerns the observation by his Honour: “It is up to you how you deal with the evidence and how you find your verdicts, but you may consider that it would be very difficult to find him guilty on one and not guilty on others. You either might accept the complainant as a truthful, accurate, credible witness or you do not. It is difficult to see how.”

Although the final sentence of this paragraph appears to have been cut short, the message conveyed is clear enough, namely that the jury might have a difficulty in coming to a different conclusion between the various counts charged. It was, however, a direction that watered down the requirement for the jury to consider each count separately by reference to the evidence relating to it.

Moreover, although it seems not to have had that result, (since differing verdicts were returned), it potentially risked inviting an all or nothing approach. It also potentially invited a form of tendency reasoning, whereby the jury would consider it appropriate to transfer acceptance of guilt on one count for which there was convincing evidence, to a finding of guilt upon the other counts where the evidence was not so clear, or where a residual doubt may have arisen.

Of more immediate relevance, however, is the submission that the directions were inadequate in not including the instruction propounded by Dunford J in RAT , noted earlier. So far as that submission depends upon the need for a direction in the precise terms proposed by his Honour, I would reject it, since for the reasons earlier given, I am of the opinion that it overstates the law, and that Jones does not constitute authority for it.

Although not fatal to the direction, as Spigelman CJ noted in ARD 2000 NSWCCA 443 (at para 3), Barwick CJ and Owen and Walsh JJ made reference, in Lucas v The Queen (1970) 120 CLR 171, to the general undesirability of complicating a summing up by reference to extraneous considerations, such as the consequence of a particular verdict, unless legislation so requires (as it now does in cases involving special hearings for those found unfit to be tried, and in cases where an issue of mental illness arises - Mental Health (Criminal Proceedings) Act 1990 sections 21 and 37) The point at issue for this appeal concerns not this point, but rather, whether the RAT direction, or some variation of it, should have been given by the trial judge.

Caution needs to be exercised in giving any direction of that kind, lest the jury see it as an encouragement to a process of reasoning in the obverse way previously mentioned. For them to assume that a tendency or propensity to similar behaviour has been established by proof of some counts, which might then make it more likely that the accused offended similarly in relation to the remaining counts, would involve a prohibited line of reasoning.

On the other hand, there is merit, as a matter of common sense, in reminding the jury that the existence of a question mark in their minds concerning the credibility, or reliability of the evidence given by a complainant or central witness in relation to one count, may properly be taken into account, in conjunction with all the other circumstances of the case, when they consider the reliability or credibility of the evidence of that complainant or witness in relation to the other counts.

An additional consideration arises so far as the giving of a direction of this kind may alleviate the concern which might otherwise subsequently arise, where differing verdicts are returned in relation to a multi count indictment. In such a case, the Court may feel greater confidence than the jury has focussed closely upon the issues of reliability and credibility of the evidence going to each charge, and has found good cause for differentiating between the individual counts.

In Robinson (2000) 111 A Crim R 388, Fitzgerald JA said in relation to such a direction, (at para 9): “The circumstances might be such that the jury should also be instructed that if they hold a reasonable doubt concerning the reliability of a complainant’s evidence on one or more counts, whether by reference to the complainant’s demeanour or for any other reason, they must take that into account in assessing the reliability of his or her evidence in relation to other counts.”

I have two reservations in relation to this direction: first, that it appears to be dependent upon the jury having a reasonable doubt concerning the reliability of the complainant’s evidence, on any one or more of the counts; and secondly, that if given, it is to be phrased in mandatory terms.

I see no reason to confine the direction to reliability. It is equally, if not more, applicable in relation to credibility or truthfulness. Moreover, I see no reason to do more than bring to the attention of the jury the circumstance that, depending upon the nature of, and the degree of, any question which arises in relation to the evidence going to the count in respect of which they have a reasonable doubt, it is both permissible and proper for them to take that into account when considering the evidence of the complainant or central witness upon the remaining counts.

This should, however, be kept in context, in that the jury should be reminded that it is but one of the circumstances that they can properly take into account, when evaluating the overall credibility and reliability of the complainant, as it impacts upon the remaining counts, including any other fact or circumstance which may provide an explanation for any question which may arise. It is, in my view, inappropriate for any direction, or comment, in this regard to become elevated, by reason of the mandatory terms in which it is expressed, to a principle of law or to operate as a straight jacket for the jury.

In some cases, where it is obvious that the witness’s reliability or credibility has been seriously undermined, in relation to one count, a strong comment may be appropriate. In other cases, it may be appropriate to indicate to the jury that they may have a difficulty in accepting the witness’s evidence on some counts, or on particular counts, if they have a reasonable doubt about his or her reliability in relation to the other counts. In other cases a more neutral reminder of the entitlement of the jury to take such matter into account may suffice. In yet other cases, it may not be necessary to say anything.

In this regard, I similarly share the concern expressed by McHugh J in KRM v The Queen (2000) 75 ALJR 550, and by Spigelman CJ in this appeal, as to the danger of adding ever more directions and warnings to the repertoire of a trial Judge. It is not to be forgotten that the purpose of a summing up is not to deliver an impeccable and exhaustive lecture on the law, or even upon all matters that might conceivably impact on the fact finding process. Nor is it to state the obvious. Rather, it is to explain the relevant law in the light of the issues which arise, and to give the jury such assistance as they need in determining those issues upon the facts relevant to them.

Where it is necessary that a direction of the kind discussed be given to ensure a balance of fairness, then it should be phrased in terms that meet the needs of the case, and with the reminder that the assessment of reliability and credibility remains entirely a matter for the jury. Whatever is said, however, it is critical to ensure that it is not so worded as to encourage or invite the obverse form of reasoning which derives from tendency or propensity.

In my view, the somewhat brief direction given in this case insufficiently brought these matters to attention. Once his Honour opened up this subject, it was necessary for him to give a sufficient direction drawing to attention the considerations mentioned.

When taken in conjunction with the absence of a Crofts direction, I have reached the conclusion, although not without some hesitation, that the appellant lost the opportunity for a fair trial

I would accordingly allow the appeal, and order a new trial.

GROVE J : I have had the advantage of reading the draft judgments of Spigelman CJ and Simpson J and I gratefully adopt their respective expositions of the facts and background material relevant to the appeal. By reference to that material I can express my own views with relative brevity. Since the preparation hereof, I have also read the judgment of Wood CJ at CL.

The grounds of appeal relied upon by the appellant were as follows: (1) The verdicts of guilty in relation to counts 1 to 5 are unreasonable and cannot be supported in light of the verdict of not guilty in relation to ground 6. (2) The trial miscarried due to the failure of the learned trial judge to direct the jury that the delay in complaint, or failure of the complainant to make a complaint was a matter to be taken into account in its assessment of the credibility of the complainant. (3) The trial miscarried as a result of the failure of the learned trial judge to direct the jury, once a question on the subject was asked in the course of its deliberation, that it was not for the appellant to establish a motive in the complainant to lie or that a failure to establish a ‘reason for coming forward and report the evidence that occurred some twenty years ago’ assisted or supported the case against the appellant and that such a reasoning process was erroneous and inverted or reversed the onus of proof .

I agree with the reasons of Spigelman CJ for rejecting ground 1 and in particular with his qualifications upon the authority of cases such as R v RAT (2000) 111 A Crim R 360, R v Barnett [2000] NSWCCA 283, R v Carbone [2000] NSWCCA 387, R v Dixon [2001] NSWCCA 39 and R v S [2001] NSWCCA 204 in their purported applications of Jones v The Queen (1997) 191 CLR 439. I would add my concurrence to that of Wood CJ at CL with Spigelman CJ concerning that case and express my agreement with observations which he has made.

In relation to grounds 2 and 3 the draft judgments which I have mentioned recite extracts of directions given by the learned trial judge and I will not set them out again.

The question by the jury and the presiding judge’s response which are the subjects of ground 3 are set out in the judgment of Spigelman CJ at pars 167 and 168. I am unpersuaded of either proposition advanced on behalf of the appellant that the jury’s question implied that inquiry was being directed to motive of the complainant to lie or that his Honour’s response invited reversal of the onus of proof. I would reject this ground.

Ground 2 directly complains of the absence of a “balancing” direction in favour of the appellant to the effect that delay in complaint may be taken into account in assessment of the complainant’s credibility when giving a direction pursuant to s107 of the Criminal Procedure Act 1986 drawing attention to the possibility of reasons for such delay: Crofts v The Queen (1996) 186 CLR 427. The firstmentioned direction has been said to be required as a general rule: Kilby v The Queen (1973) 129 CLR 465. In this trial the learned trial judge adverted to reasons testified to by the complainant but did not in express language associate delay in complaining and credibility.

The warnings which he gave to the jury accorded with the requirements Longman v The Queen (1989) 168 CLR 79. Whilst I respect the views expressed by Simpson J concerning Crampton v The Queen 2000 75 ALJR 133 I do not share her conclusion that his Honour’s warnings were inadequate in drawing attention to the impact of delay upon the appellant’s ability to respond to matters of fact. No complaint was made at trial (which preceded delivery of judgment in Crampton ) and although detached from the atmosphere of the trial it is compelling to infer that the factual contest including the effects of lapse of time upon the abilities of all concerned to engage in it must have been obvious to the jury.

The issue is whether ground 2 must be sustained. In R v Davies & Ors 1985 3 NSWLR 276 it was affirmed (applying Kilby ) that a failure to make timely complaint should be the subject of a direction that such circumstance was relevant to the assessment of a complainant’s credit. However it was concluded in that case that the jury could not have been under any misapprehension as to the relevance to credit of the issue of complaint and leave to rely upon a ground similar to the one under consideration was refused. The requirement for direction articulated in Davies was approved in Crofts but no criticism was made of the course adopted to dispose of the appeal adversely to the appellant. I consider that the present circumstances and the content of directions in essence parallel those in Davies and I would reject ground 2.

Spigelman CJ has raised the question of a particular direction in relation to multiple counts such as suggested by Fitzgerald JA in R v Robinson (2000) 111 A Crim R 388: “The circumstances might be such that the jury should also be instructed that if they hold a reasonable doubt concerning the reliability of a complainant’s evidence on one or more counts, whether by reference to the complainant, or for any other reason they must take that into account in assessing the reliability of his or her evidence in relation to other counts.”

The Chief Justice made explicit his “considerable reluctance to add to the number of directions and warnings which trial judges are already required to give” and his consciousness of the adjuration of McHugh J in KRM v The Queen (2000) 25 ALJR 550: “The more directions and warnings juries are given the more likely it is that they will forget or misinterpret some directions or warnings”.

Nevertheless he concluded that it is desirable to supplement the direction as to treating each count separately in a word against word case and that reference ought to be made to the effect on credibility if the jury found itself unable to accept the complainant’s evidence on any one count. I recognize, as his Honour does, that the absence of such a direction would not necessarily be fatal but a final conclusion was: “The crucial matter is to indicate to the jury that any doubt that they may form with respect to one aspect of the complainant’s evidence ought be considered by them when assessing the overall credibility of the complainant, and, therefore, when deciding whether or not there was a reasonable doubt about the complainant’s evidence with respect to other counts.”

It is self evident that a trial judge must make a decision whether to give such an indication before it is known whether the jury will discriminate in its verdicts upon separate counts. Whilst I consider that it would be prudent to give such a direction in some cases I am apprehensive that language incorporating the concept of cruciality will be interpreted as mandatory. For my part I would refrain from expression in any terms which might be so interpreted. There may be cases in which it could be necessary to assist a jury by way of an instruction such as is under discussion but in some way it may not. In my view it would suffice to commend the matter for consideration of trial judges.

The other members of the Court have concluded that the convictions should be quashed. Spigelman CJ (Carruthers AJ agreeing) and Wood CJ at CL propose that a new trial be ordered. For the reasons already expressed I do not share the conclusions of Simpson J.

The Chief Justice reached his conclusion as to appropriate order by reason of the combined impact of the absence of directions on the effect on credibility of delay in complaint and of the acquittal on one count. As I am of opinion that the acquittal on the single count did not reveal any flaw in the jury’s discharge of its function I would not consider it relevant to the integrity of the verdicts on counts 1 to 5, either standing alone or in combination.

With every respect to the contrary conclusions reached by Spigelman CJ, Carruthers AJ and Wood CJ at CL, I am unpersuaded that the directions and responses of the learned trial judge were inadequate to the circumstances. The first instance judge was in an advantaged position to assess the extent of need for assistance to the jury in the atmosphere of the issues at trial and the manner in which they were fought with the particular reference to the addresses of counsel. The latter were also in a position to seek additional explicit directions if such were thought necessary or desirable. I do not consider that miscarriage is to be perceived.

I would dismiss the appeal.

SIMPSON J: On 15 August 2000 the appellant was arraigned in the District Court on an indictment containing six counts to which he entered a plea of not guilty. A jury was empanelled and on 23 August returned verdicts of guilty on counts 1 to 5 inclusive and not guilty on count 6. The appellant now appeals against the convictions. He has abandoned an application for leave to appeal against the sentences subsequently imposed.

Each of the charges was of sexual misconduct involving the same complainant, BP. Counts 1 - 4 alleged indecent assault on a person under the age of sixteen years. Counts 5 and 6 alleged sexual intercourse without consent with a person under the age of sixteen years.

The Crown case was that the offences had been committed over a period of just over one year, beginning on 29 January 1981 and ending on 4 February 1982. It depended on the evidence of the complainant who was aged eight at the date specified in the first count and nine on the dates specified in the remaining counts.

The Crown case was that, at relevant times, the complainant lived with her mother in a home unit in Warren Road, Marrickville. Her grandmother also lived there part of the time but spent some days each week with her husband in the Wollongong area. The appellant lived with his wife and four year old son Jimi in a neighbouring unit in the same block. The complainant was (and is) profoundly deaf. She gave her evidence through an Auslan (sign language for the deaf) interpreter. She attended a special school for deaf children, to and from which she was transported by taxi.

Her mother was in employment. On two days each fortnight her employment prevented her from being at home when the complainant returned from school. She therefore made arrangements for the complainant to be cared for during those afternoons by the appellant’s wife. It was the complainant’s practice, on these occasions, to go to the appellant’s home after school, where she played with Jimi until her mother collected her.

The complainant gave evidence of each of the six offences alleged. She said the first occurred on 29 January 1981. She was able to pinpoint this date as the second day of school term after summer holidays. She said she was at the appellant’s home in the care of his wife, and playing a game of hide and seek with Jimi. The appellant returned home from work and joined in the game. He directed the complainant to go to the bedroom he and his wife occupied, which she did. He followed her, unzipped her school uniform, and fondled her breasts. Jimi came looking for them and the appellant re-zipped the complainant’s school uniform.

The complainant said the second offence occurred on or about 31 May 1981, a few days after her ninth birthday. She had accompanied her mother and grandmother on a visit to the appellant’s apartment but had returned to her own home with her grandmother. She went onto the balcony from where she saw the appellant signalling her to come back to his unit. She described the expression on his face as “quite horrible”. She did return to his apartment, and he told her that she, he and Jimi were again to play hide and seek. When she went looking for Jimi she encountered the appellant behind his bedroom door. He again touched her breasts, then unzipped his pants, exposed his penis and used the complainant’s hand to masturbate himself.

The complainant’s evidence was that the third offence was committed on 5 June 1981, again at the appellant’s home. On this occasion, she said, she had gone to use the toilet when the appellant entered and shut the door. She was embarrassed and stood up to dress herself. The appellant grabbed her arm and held her hand on the outside of his clothes, rubbing it up and down against his penis. They left the toilet and the appellant directed the complainant to his bedroom where he pushed his penis into her mouth. He then signalled to her that if she revealed what had happened he would cut her throat.

The complainant said that the offence that constituted count 4 was committed on 18 June 1981. Again she was spending the afternoon at the appellant’s home. Jimi was occupied with his own activities and the complainant was on the balcony. The appellant rubbed her in the region of the vagina, unzipped and lowered his pants and exposed his penis. Although it was flaccid he attempted, using his finger, to push it into her vagina.

The complainant’s evidence was that the offence constituting count 5 was committed on or about 24 July 1981. She could pinpoint this date because it was close to the birthday of a friend. The complainant was again at the appellant’s home, playing with Jimi. The appellant told her to hide in the bedroom. She did so. While she was there the appellant pushed her head onto his penis and inserted his penis into her mouth, together with his little finger. She felt his penis become erect in her mouth. He then ejaculated.

The complainant’s evidence was that the offence constituting the final count was committed on 4 February 1982. This was the first day of the new school term. The complainant was at the appellant’s home as usual. The appellant arrived home from work and invited her to play hide and seek. He directed her to the bedroom door, pulled down his pants and exposed his penis. The complainant attempted to resist but the appellant pulled her pants down and inserted his penis into her vagina and also into her anus. The complainant attempted to leave the bedroom to go into the lounge room but the appellant told her to get under the bed. She did so. She was around the side of the bed, her head under the bed with her legs emerging from underneath it. She kept them straight and close together but the appellant pulled them apart, pulled her pants down, licked her on the vagina and inserted his finger into the vagina. The complainant was pushing him on the head and telling him to stop. When he ceased she came out from under the bed and dressed herself.

The complainant also gave evidence of other incidents, not the subject of any specific charge. She said that the appellant regularly touched or bit her on the breasts. Of particular significance is an event she described as having taken place at about the end of February 1981. On this occasion, the complainant said, the appellant had sent her to the bedroom during a game of hide and seek, and then told her to get under the bed. This was the same bed as that that had featured in the incident the subject of count 6. The complainant described the bed as just a mattress, very low to the floor. Initially, she said, she refused, but the appellant eventually grabbed her arm and told her again to get under the bed, which she did. She said her body was under the bed but her legs were protruding, and that there was only about one and a half or two inches between her face and the bottom of the mattress. She said she was at first lying face down but the appellant turned her over so that she was lying on her back. The appellant pulled up her dress, pulled her pants across, and then put his finger in her vagina, scratching her with his fingernail. • • •

The complainant made no report of any of these incidents to anybody until October 1998 when she attended the Marrickville Police Station. She said that she had not told her mother because her mother often smacked her and she had not told the appellant’s wife, who was present in the apartment, usually in the kitchen, on the occasion of each incident, because she was afraid. In giving this evidence she made specific reference to the throat slitting gesture she said the appellant had made in June 1981, on the occasion of the events the subject of count 3. She also said that she had not said anything to the appellant’s wife because her ability to communicate was limited as a result of her deafness. She said, however, that she had signalled to the appellant’s wife, asking where her mother was, and that she had told her mother that she did not want to go to the appellant’s home, but that her mother had insisted she do so.

The complainant’s mother, who also gave evidence, denied that she frequently smacked the complainant, but she confirmed that, at about the age of eight or nine, the complainant had said she did not want to go to the appellant’s home.

The complainant was the only witness in the Crown case who gave evidence relating to the actual incidents alleged. Her evidence as to these matters was entirely uncorroborated. Her mother also gave evidence, as did three police officers. It will be necessary to return to some aspects of their testimony.

The appellant was arrested and interviewed on 19 May 1999. There is no suggestion in the evidence that he had any earlier notice of the complainant’s allegations. The defence case

The defence case was a denial of each allegation. The complainant was cross-examined extensively. Cross-examination focussed on a number of matters. One was the size of the appellant’s apartment, which was admittedly very small. The point of this cross-examination was to lay the foundation for suggesting the unlikelihood of the appellant’s having behaved as the complainant alleged, in a small apartment, in which his wife and son were present. It was further put that it would not have been feasible for the conduct described to have taken place without having been observed by the appellant’s wife or son. There was evidence of the size and layout of the apartment by way of photographs and sketch plans.

Another matter of significance about which the complainant was cross-examined concerned the bed under which she said two of the assaults (one not the subject of specific charge, the other the subject of count 6) had been committed. This is a major issue in the appeal.

The appellant gave sworn evidence. He gave his evidence through a Macedonian interpreter, although he had a tendency to lapse into heavily accented English. This caused some transcription difficulties and also some apparent difficulties in understanding exactly what the appellant was saying. The substance of his evidence was to deny each allegation.

It may here be observed that the appellant was considerably handicapped in presenting a factual defence by reason of the lapse of time - almost eighteen years in the case of the earliest allegations - between the dates of the events in question and the making of the allegations. Plainly, although the complainant was quite specific as to dates, it was extremely unlikely that the respondent would be able to produce any material that would contradict the complainant’s assertions by way of alibi, or other evidence casting doubt upon aspects of her claims.

He denied that he had ever played games with the complainant and his son, saying that he was too tired after a long day’s work to do so. He agreed that the complainant and his son were in the habit of playing hide and seek. In this evidence he was supported by both his wife and his son, who was twenty-four by the time he gave evidence.

The main thrust of the defence case was directed to the complainant’s allegations involving the bed in the appellant’s bedroom. He sought to make a positive case that the incidents described by the complainant could not have occurred. Of most importance in this respect was the evidence given by and on behalf of the appellant concerning the bedroom furniture in the apartment. This evidence was very hard to follow, and drew an observation to that effect from the trial judge. The transcript shows that the evidence given by the appellant was quite confused. The essence of his evidence appears to be this. He and his wife had moved into the Warren Road apartment in 1975. It was an apartment with two bedrooms. In 1981 he and his wife occupied one bedroom, Jimi the other. (Jimi was born in April 1976.)

At first the appellant said that in 1981 he and his wife slept in a double or queen sized bed that they had purchased in 1975. Subsequently he seemed to say that, when Jimi was about three or four, he and his wife had purchased a new queen sized bed, and had given their original bed to Jimi. He said that both beds were of the same type, an ensemble consisting of a padded bed base topped by a mattress. The bottom of each bed was about two and a half or three inches from the floor.

By the date of the trial the appellant and his family had moved to a three bedroomed apartment in Hill Street, Marrickville. According to their evidence, they had taken with them, and still had, the same two beds. Jimi was still living with his parents. Photographs of the two beds in the Hill Street apartment were in evidence. They were taken on 19 May 1999 by police who investigated the complainant’s allegations. They became Ex E in the trial.

The photographs show beyond doubt that, if these were the beds in the Warren Road apartment, the complainant’s account of count 6 could not be accepted. Nor could she be accepted in relation to the uncharged earlier incident involving the bed. One principal issue for the jury, and for this court, was whether the Crown had established beyond reasonable doubt that neither of the beds in Ex E was the bed in the appellant’s bedroom in 1981 and 1982. If it had not, it could not succeed on that count.

The appellant’s evidence was supported by his wife. She said that the queen sized bed was bought in 1980, and that was when the double bed was given to Jimi. Although in examination in chief and cross-examination of the appellant and his wife a great deal of attention was paid to when the new bed was bought, that hardly matters. If the two beds shown in Ex E were the beds in the apartment at the relevant time then the complainant’s story could not have been true. Indeed, it is accurate to say that if the jury thought there was any reasonable possibility that the two beds shown in Ex E were the beds in the apartment at the relevant time, then they had to have a doubt about the complainant’s evidence on that count. The consequences of such a finding are those stated in R v Jones (1997) 191 CLR 439. That is, the doubt would have to be translated to a doubt about the complainant’s reliability in general, and the appellant would be entitled to an acquittal.

The photographs in Ex E were shown to the complainant. She denied that the beds shown in the photographs were the beds in the apartment. She said that an ensemble bed, of the kind shown in the photographs, was in Jimi’s room, but that the bed in the appellant’s bedroom was a bed of a different kind, without a bed base. She said she could not recall what sort of legs the bed had because it was covered by blankets.

It is of very considerable significance that it was on the count, and only on the count, that involved allegations that the sexual misconduct had taken place under the bed, that the appellant was acquitted. The only possible inference is that the jury were not satisfied beyond reasonable doubt that the events described by the complainant had occurred. In other words, they had a reasonable doubt about the reliability of her evidence in relation to this allegation.

On behalf of the Crown it was suggested that the explanation for the acquittal on this count might lie in a confusion that emerged about the date of the event alleged. It is necessary to refer to some of the specific evidence the complainant gave. In her evidence in chief she was asked: “Was 4 February 1982 the first day of the new school term?” She assented to this proposition. No further attention was paid, in examination in chief, to the date of the incident. The complainant was extensively cross-examined about the bed which featured in her evidence of this incident. Towards the end of a long cross-examination defence counsel asked her about the date of the incident. She accepted that in a statement to police she had said: “I remember another time when I was at Tommy’s house. It was Thursday 4 February 1992 this was the first day of the new school term.”

She said, however, that the school term that year started on a Wednesday, and that she did not go to the appellant’s home on a Wednesday but on Thursday. By clear implication, therefore, she was claiming in her evidence that the incident occurred on the second day of term, not the first. This was different to what she had said to police, and different to what she accepted in the questions asked by the Crown Prosecutor.

The Crown’s submission was that the jury might have found itself unable to accept that this charge had been proven beyond reasonable doubt because of the discrepancy in the complainant’s description of when it had occurred.

I am unable to accept this as a reasonable possibility. The discrepancy is minor. Whether the event occurred on the first or second day of the school term had no bearing on whether the complainant’s evidence in this regard should be believed. More significantly, no mention was made in the summing up of any argument put by defence counsel to the effect that the discrepancy as to the date of the event seriously affected the probability that it had occurred. The addresses of counsel were not put before this court, and were probably not recorded, but the absence of any reference in the summing-up to such an argument, and the absence of any request for a direction in this regard, suggest that, despite the cross-examination, the date never emerged as an issue. In my view, the uncertainty in the complainant’s evidence as to the timing of the event the subject of count 6 cannot explain the different verdict on that count.

Part but not all of the significance to be attributed to the acquittal on this count derives from the fact that it was only in relation to this count that the appellant was able to adduce any positive evidence. In this respect it is obvious that the lapse of eighteen years between the time the offences were alleged to have been committed and the first notice the appellant was given of the allegations at least potentially prejudiced his capacity to answer them. The appeal

The principal ground of appeal advanced by the appellant was made in reliance upon the decision of the High Court in Jones. That case, in my view, cannot relevantly be distinguished from the present. There the complainant alleged that the appellant had sexually assaulted her on three occasions. She had made no complaint for four years. Her evidence as to the acts was uncorroborated.

The majority in the High Court said: “The jury’s finding of not guilty on the second count damaged the credibility of the complainant with respect to all counts in the indictment. Implicit in the appellant’s acquittal on the second count was a rejection of the complainant’s account of the events which were said to give rise to that count. The jury may have acquitted because the unshaken evidence of the appellant’s wife and daughter with respect to the absence of opportunity engendered a reasonable doubt about the appellant’s guilt on the second count. They may possibly have acquitted because they were of the view that the complainant had fabricated the circumstances given rise to the second count. Having regard to the convictions on the first and third counts, the latter alternative seems an unlikely explanation of the verdict on the second count. Whatever the explanation may be, however, the jury's rejection of the complainant’s account on the second count diminished her overall credibility. The only reasonable conclusion is that the jury were not satisfied beyond reasonable doubt of the truth of her evidence concerning the incident the subject of the second count. Having regard to the terms of the second count, we do not think that the complainant’s confusion about whether the offence took place on a Thursday or a Saturday could explain the acquittal on that count. If that was the reason the jury rejected her evidence, they would also have acquitted the appellant on the first count. It is difficult then to see how it was open to the jury to be convinced beyond a reasonable doubt of the guilt of the appellant with respect to the first and third counts. There is nothing in the complainant’s evidence or the surrounding circumstances which gives any ground for supposing that her evidence was more reliable in relation to those counts than it was in relation to the second count. Moreover, two other factors made it necessary for a reasonable jury to scrutinise the complainant’s evidence with considerable care - (1) her delay in making the complaint; and (2) the lack of any corroborative evidence, in particular, the absence of any medical evidence.” (p 453, internal references omitted.). At p455 their Honours said: “In our opinion, the proper application of the test formulated by the majority of this Court in M [(1994) 181 CLR 487] required the convictions of the appellant to be set aside on the basis that they were unsafe and unsatisfactory. Given the jury’s finding on the second count, it was not open to them, on the whole of the evidence, to be satisfied beyond reasonable doubt of the guilt of the appellant on the first and third counts. Once the jury found that the evidence of the complainant with respect to the second count lacked sufficient cogency to convict, the Crown case on the first and third counts wore a different complexion. For it meant that, when her evidence could be set against other reliable evidence, it failed to carry sufficient conviction to reach the criminal standard of proof. As we have already said, nothing in the complainant’s evidence gave any ground for thinking that the quality of her evidence was higher in respect of the first and third count than it was in respect of the second count. When the credibility factor is combined with the uncorroborated nature of the complaints and the effect of the lengthy and unexplained delay in the making of the complaints, the convictions on the first and third counts can only be regarded as unsafe and unsatisfactory.”

These observations apply with equal force to the present case.

R v James [1999] NSWCCA 191 was a majority decision of this court. The case had significant parallels with the present. There the appellant had been charged with five counts of indecent assault. He was convicted of three, acquitted of one, and the jury was unable to agree on one. Although, almost a year after the date on which the last offence was alleged to have been committed, a diary kept by the complainant had been found in which statements consistent with her later allegations appeared, the complainant made no report or complaint until (coincidentally) eighteen years later. She was thirteen and fourteen at the times the offences were alleged to have been committed. In that case, like the present, the appellant had borne the obvious disability in the presentation of a factual defence by the lengthy lapse of time. He was, however, like the present appellant, able to adduce evidence that one of the incidents described by the complainant simply could not have occurred as described by her. This, in that case, was because he was able to call evidence that a building in which the complainant claimed one of the offences had been committed simply did not exist in the relevant location. It was on the count related to this allegation that the jury acquitted. The jury plainly had a doubt about the reliability of the complainant’s evidence on this count. There being nothing in the evidence to show that the complainant’s evidence on the other counts was more reliable, the doubt had to affect her evidence on both counts also.

The same reasoning must apply in this case. Once it could be seen that the jury experienced a doubt about the complainant’s credibility in relation to the sixth count, the count on which the appellant was able to produce some positive evidence in response to the complainant’s assertions, her reliability suffered an irremediable blow. The circumstances did not permit the jury reasonably to doubt her credibility or reliability on the one count which the appellant was in a position to answer, and not the other counts.

On the Jones principles alone, in my view, the appeal must succeed. For this reason, I propose that the appeal be allowed, the convictions quashed and a judgment of acquittal entered.

Since dictating the above, I have had the opportunity of reading in draft the judgments of Spigelman CJ, Wood CJ at CL and Grove J. I do not in any respect dissent from the exposition of the law, nor the analysis of the decision in Jones and subsequent cases in which that decision has been considered or applied, undertaken by the Chief Justice and further explored by Wood CJ at CL. Those analyses have caused me to consider again the approach I have taken above.

Applying those analyses to the facts of this case, I remain of the view that the question as to the complainant’s reliability created by reason of the acquittal on the last count is necessarily so significant as to require its translation to the other counts. The most substantial factor that leads me to this conclusion is the length of time between the dates of the alleged offences and the charging of the appellant. This put the appellant at a substantial disadvantage. It is of considerable significance that it was only in relation to the single charge in which he was able to adduce positive evidence that he was acquitted.

Notwithstanding the qualification which, by reason of the expositions of the Chief Justice and Wood CJ at CL, are now placed upon the previous approach to the application of Jones , I remain of the view that the acquittal on the last count entitles the appellant to an acquittal on each of the remaining counts.

Since other members of the bench take a different view on the principal ground of appeal, it is necessary to deal with two remaining grounds. These were expressed as follows: Ground 2 The trial miscarried due to the failure of the learned trial judge to direct the jury that the delay in complaint, or failure of the complainant to make a complaint, was a matter to be taken into account in its assessment of the credibility of the complainant. Ground 3 The trial miscarried as a result of the failure of the learned trial judge to direct the jury, once a question on the subject was asked in the course of its deliberations, that it was not for the appellant to establish a motive in the complainant to lie or that a failure to establish a ‘reason for coming forward and report the evidence that occurred some twenty years ago’ assisted or supported the case against the appellant and that such a reasoning process was erroneous and inverted or reversed the onus of proof. Ground 2

So far as I am able to ascertain from the evidence, the complainant first made her allegations to police in or about October 1998. The appellant was arrested on 19 May 1999. There is nothing to suggest that he had previously been made aware of the complainant’s allegations. His first awareness of what he had to answer, therefore, came more than seventeen years after the date the last incident was alleged to have occurred. The handicap the delay imposed on his capacity to defend the charges is obvious.

In relation to delay the judge directed the jury in the following terms: “There is another direction that I will give you, a legal direction, and that is this: that it is about nineteen years - eighteen, nineteen years - since these events are alleged to have taken place. The result of that is that it can make it very difficult for an accused person to adequately test the evidence. There has been criticism, for example, in this case that the police did not even get the manufacturer’s name of that Queen bed; if that sort of material had been looked at in due time perhaps an accurate date of manufacture could be worked out and that could be very material. It is quite plain, I think, if these ensembles, that is a mattress on top of a padded base were in use as at the time of these alleged events, then it is almost impossible to get under it; it is not even clear that someone’s head could or could not go under them. That is the sort of thing that does come into play when there has been a very great delay. If you are of the view it was difficult to test the complainant’s evidence because of the passage of almost twenty years, then you are entitled to take the view that it would be dangerous to convict on that evidence alone unless, scrutinising the evidence with great care considering the circumstances relevant to its evaluation and paying heed to the warning I have given you, you are satisfied of its truth and accuracy. That is really a draft of what I have told you already about the scrutiny of the complainant’s evidence.”

A little later, after giving some other directions, his Honour said: “There is a legal direction I should have given you; I will do that now. There has been some reference to the fact that she did not complain when Mr Young put to her that she never attempted to say anything to Mrs Markuleski. She said she was afraid. The law requires me to give you a direction that an absence of complaint in respect of the commission of an alleged offence by the person on whom the offence is alleged to have been committed, or delay by the complainant in making such a complaint, does not necessarily indicate that the allegation was false. I am directed to tell you that there may be good reason why the victim of a sexual assault would delay in making or refrain from making a complaint about the assault. The complaint, presumably, was made about October 1998, presumably before or at the time of the first police statement. You remember the evidence the complainant gave about it. Firstly, she said she was frightened of being smacked by her mum and also she was frightened because of the accused saying ‘shush’ and making the motion across the throat sign. That was the reason she gave.”

This last direction was required by s 107 of the Criminal Procedure Act 1986, and is not challenged. The complaint now made is that it was not “counterbalanced” by a direction drawing attention to the impact of the delay on the complainant’s credibility.

That delay in complaint is relevant to the jury’s assessment of the credibility of the complainant, and that a jury should, ordinarily, be expressly be so directed, was established in the decision of the High Court in R v Kilby (1973) 129 CLR 460. The decision in Kilby pre-dated the enactment of s 405B of the Crimes Act 1900 (the forerunner of s 107 of the Criminal Procedure Act) but was unaffected by that enactment: R v Davies (1985) 3 NSWLR 276. Davies was, in turn, approved in Crofts v R (1996) 186 CLR 427. As I read these decisions, the focus of the direction they required was on the credibility of the complainant in the light of a failure to make an early complaint.

A more recent development in the approach of the High Court to the directions to be given in relation to delay in complaint when allegations of sexual misconduct are made is to be found in R v Crampton [2000] HCA 60; 75 ALJR 133. There the High Court, adopting what had earlier been said in R v Longman (1989) 168 CLR 79, held that what is required is an explicit direction, not only as to the impact of delay of itself on the complainant’s credibility, but drawing the attention of the jury to the impact of delay on an accused’s ability to make a factual response to the complainant’s allegations, and to challenge detail or factual matters included in the complainant’s evidence, so as to raise a doubt about the accuracy of the prosecution case. What is required by Crampton goes beyond what is required by Kilby , or Crofts , or Davies ; the focus is not simply upon whether the mere fact of delay affects the complainant’s credibility; it is upon the practical difficulties that that creates for the accused. An obvious example of what the High Court had in mind (in Longman as well as in Crampton ) would be alibi evidence that might be completely lost after a substantial lapse of time. No doubt there would, in individual cases, be particular matters that should be drawn to the attention of the jury. The precise direction required will vary from case to case, depending upon the factual matters raised.

Indeed, the present case presents a very good illustration of the need for such a direction. The appellant was able to produce evidence sufficient to cast a doubt upon the complainant’s description of the events the subject of the sixth count, and this resulted in his acquittal on that count. In relation to some of the other counts, the complainant had been quite specific as to dates and other circumstances surrounding the events she described. But it could hardly be expected that, after seventeen years, the appellant would be able to produce any concrete response to that evidence. This is precisely the kind of difficulty that should, in compliance with Longman and Crampton, be drawn to the jury’s attention.

I am persuaded that the directions given were not sufficient and that this ground should succeed. Standing alone, it would entitle the appellant to a new trial. Ground 3

It appears that, immediately upon their retirement, the jury requested access to various documents that had been referred to in evidence but not tendered as part of the evidence. As read onto the transcript, their note continued: “We would also like to propose a question to [the complainant] as to the reason for coming forward and reporting the evidence that occurred some twenty years ago.”

After discussion with counsel, his Honour responded to that part of the jury request in the following terms: “As to your final enquiry that you would like to propose a question to [the complainant] as to the reason for coming forward and reporting an incident that occurred twenty years ago, I have given you a bit of a direction that we are required to go in in that respect. Again, the difficulty is that [the complainant] is not here and the whole of the evidence in the case is finished. We are, in effect, stuck with that situation. The Crown is not in a position to call her back and certainly would not want to call her and it is not for me to put her back in the witness box after the case is finished insofar as the evidence is concerned. So I am sorry to be so negative, members of the jury, but that is about all I can say.”

In the past it has not infrequently happened that the Crown has, either in cross-examination of the accused, or in final address, posed the rhetorical question “why would the complainant lie?”. Whether expressly raised by the Crown or not that is a question that “hovers over cases of this nature” ( R v F (1995) 83 A Crim R 502). It is now firmly established that such a question should not be asked of an accused who gives evidence, nor posited to the jury by way of submission by the Crown: Palmer v R (1998) 193 CLR 1; R v F (above); R v E (1996) 39 NSWLR 450; R v Jovanovic (1997) 42 NSWLR 502. The reasons that it is so are fully explained in the cases cited.

That, of course, is not precisely what happened here. Firstly, it was the jury who, of its own volition, raised the question.. Secondly, the question was not expressly framed in terms of “why would the complainant lie?”

Indeed, the question is not easy to interpret. On its face it conveys a degree of scepticism about the complainant’s credibility which must have operated to the appellant’s advantage. However, it also indicates that the jury was interested in exploring the reasons for her delay in coming forward. There was a danger that, unless given an explicit direction, the jury would reason that, in the absence of evidence of a motive in the complainant to lie, her credibility was enhanced. It therefore became necessary to deal with a matter that was obviously of concern to the jury, and potentially leading them on a false train of enquiry. In these circumstances, the response should have recognised the potential for an impermissible reasoning process, and headed off that danger. The jury should have been told that , while evidence of a motive in the complainant to lie could be a very material factor in assessing her credibility, the absence of evidence of such a motive (as distinct from established absence of such a motive) was neutral, and did nothing to advance the Crown case. (A positive claim of the absence of a motive to lie would be quite different, but, apart from a complainant’s own assertions, could rarely be the subject of evidence.)

The direction that should have been given was to the effect that, firstly, there was no evidence of the complainant’s reasons for the delay in coming forward, and, secondly, that they ought not speculate about what those reasons might be. Most importantly, they should have been told that the absence of evidence of a motive to concoct a false story about the appellant did not enhance the complainant’s credibility or strengthen the Crown case against the appellant.

I have hesitated over the question of whether this ground of appeal, alone, would have warranted the quashing of the conviction and the ordering of a new trial. However, having concluded that there was a danger of impermissible reasoning, I must also conclude that the absence of an appropriate direction gave rise to the real possibility that a miscarriage of justice has occurred. Accordingly, on this ground also I would quash the conviction and order a new trial.

The orders I propose are that, on each count on which the appellant was convicted, the conviction be set aside and a judgment of acquittal entered.

CARRUTHERS AJ: I have had the benefit of reading the judgment of the Chief Justice in draft form. I am in complete agreement with the orders which he proposes and his reasons therefor.