However, having regard to the way in which this issue was fought at trial and the evidentiary context in which this DNA material was adduced, I do not consider that any miscarriage of the trial occurred.

The eighth ground of appeal concerns the application of s38 of the Evidence Act 1995 .

The relevant evidence is helpfully set out in the judgment of James J and I need not repeat it. It is clear from the course of evidence that both the Crown and defence counsel took a tactical approach to this evidence. The Crown was, of course, aware, that the witness was apparently unable to state categorically, from a positive recollection, who opened the door on the crucial occasion. He had said that he had no memory of seeing the accused at the time, but this was not in its own terms a direct statement at all that the accused had not been there. When it is noted that immediately preceding this statement he had said that he could not remember if it was the accused or Baca-flor who opened the door, this point is emphatically established, to my mind. It was not to be reasonably expected that his memory would have improved in the interim and if the expression had become positive, it was most unlikely to be reliable. The appropriate question for the Crown to ask, on such a crucial matter, was whether the witness could now recall how he had got into the building; this was, at all events, the logically preceding matter. Having, decided to seek the unqualified answer, if he could get it, it seems to me that the Crown deliberately invited the cross-examination that occurred. It seems to me to be most unfair that the Crown should seek to take advantage of an unqualified answer that he knew was unlikely to be true upon the basis that the most he could lose was that the cross-examiner might, on the basis of the statement, bring the witness back to a state of uncertainty. In the result, however (and, no doubt on the basis of instructions) the cross-examiner sought to adduce the evidence in a way that was based upon a positive recollection, building upon the supposition that nothing out of the ordinary happened that night.

I think it would be most unfortunate if counsel, especially prosecuting counsel, thought that s38 of the Evidence Act might be used a device to nullify the effect of a cross-examination where they have themselves proceeded or permitted the evidence to proceed upon a false or unreliable basis. I agree that, in the peculiar circumstances of this case, the learned trial judge has not been demonstrated to have erred in the exercise of his discretion in permitting the further examination by the Crown prosecutor although (with unfeigned respect for his Honour) I do not say that I would necessarily have come to the same conclusion.

It should be noted that s38(1)(a) is confined to circumstances in which a witness gives "unfavourable" evidence. Other paragraphs of the subsection give other grounds for cross-examination, of which para (b) is directly applicable to the circumstances here and which would have permitted the Crown to ensure that this important evidence went to the jury upon an accurate basis. However, as I have said, he did not choose this course. It seems to me that some attention will need to be given in due course to the meaning of "unfavourable" so far as the Crown case is concerned. The Crown case is, in essence, the truth, wherever that might lead and even if it leads to a reasonable doubt about guilt. I am far from persuaded that merely because a witness declines to give evidence supporting the theory of the facts for which the prosecution contends or, indeed, gives evidence that contradicts that theory or contention, his or her evidence may thereby be regarded as "unfavourable". How can truthful evidence ever be "unfavourable" from the Crown's point of view? In the circumstances of this case, however, there were good reasons for concluding that, if not untruthful (and I do not think that this was ever suggested in the sense of being deliberate) the evidence of the witness as adduced by the cross-examination was doubtful, even if (by parity of reasoning) the evidence adduced in chief by the Crown was also doubtful.

In Penney v The Queen [1998] HCA 51 (unreported 13 August 1998) it was submitted for the appellant that the trial judge had erred in omitting to instruct the jury that, if motive was relied on by the Crown in proof of the charge (of attempted murder), then it must be proved beyond reasonable doubt. Support for this proposition was gained from R v Murphy (1985) 4 NSWLR 42 at 59-60. Penney was brought to our attention by Buddin SC, for the appellant, after argument had concluded and whilst this Court had reserved its judgment. As is clear from James J's judgment, the same argument as had been put to the High Court was put to this Court. I respectfully agree with what James J has said about it.

Penney does not advance the appellant's case. Callinan J (with whom the other judges agreed) distinguished Murphy upon the basis that the relevant portion of the summing up "was not dealing with intention or motive as such but was pointing to evidence with respect to the relationship between the appellant and his wife upon which [the jury] for an ultimate conclusion of guilt beyond reasonable doubt", citing the well known passage from the judgment of Barwick CJ in Wilson v The Queen [1970] HCA 17; (1970) 123 CLR 334. I do not read the judgment of Callinan J as endorsing the passage from Murphy.

It seems to me that, especially having regard to the explanation of Chamberlain v The Queen (No2) (1984) 153 CLR 520 expressed in Shepherd v The Queen [1990] HCA 56; (1990) 170 CLR 573, no direction along the lines submitted by counsel for the appellant was called for in this case. There may be cases depending upon circumstantial evidence in which proof of motive is an indispensable link in the chain of proof of the guilt of the accused and thus will need to be established beyond reasonable doubt. However, such cases must be rare. This case was certainly not one of them. It is always important to note that the process of reason used to move to a verdict is a matter for the jury. Thus, in some cases it may be necessary to direct them that, if they were of the view that any fact, such as motive, was, so far as they were concerned, an indispensable link in proof of guilt, then they must be satisfied of it beyond reasonable doubt before they could convict. Again, however, I do not think that this was such a case.

Accordingly, I agree with James J that the seventh ground of appeal should be rejected.

ADAMS J: I have had the advantage of reading in draft the reasons for judgment of James J. Subject to the comments below, I agree with his reasons. I agree with the orders he proposes.

The fourth ground of appeal was that his Honour, the learned trial judge "erred in failing to exclude evidence of the statistical unlikelihood of a chance match of the DNA test results of the appellant and those of the vaginal swabs of the nightdress alleged to be that of the witness Zenaida Trujillo". James J has set out comprehensively the relevant material in respect of this ground of appeal and I do not need to do so. I do not find the argument persuasive that, because it was not practicable to devise a database which took into account the probability that the semen came from a Quechua Indian, the Crown was entitled to use what it had at its command, namely databases drawn from the general population. If there was available to the jury evidence that enabled it to evaluate the possible ramifications of this omission, then it could consider the appropriate weight to be accorded to the scientific opinions that were expressed. However, in the absence of any such material, I do not see how a jury of lay persons could reasonably be expected to measure the extent of the possible impact on the statistical material that was adduced. I do not think that this highly technical evidence is a matter to which "common sense" can have any meaningful application, whether of the jury or, for that matter, the judge. The direction that they should not approach the issue "upon any strictly mathematical basis" does not, to my mind, deal with the problem: it merely ignores it.

However, having regard to the way in which this issue was fought at trial and the evidentiary context in which this DNA material was adduced, I do not consider that any miscarriage of the trial occurred.

The eighth ground of appeal concerns the application of s38 of the Evidence Act 1995 .

The relevant evidence is helpfully set out in the judgment of James J and I need not repeat it. It is clear from the course of evidence that both the Crown and defence counsel took a tactical approach to this evidence. The Crown was, of course, aware, that the witness was apparently unable to state categorically, from a positive recollection, who opened the door on the crucial occasion. He had said that he had no memory of seeing the accused at the time, but this was not in its own terms a direct statement at all that the accused had not been there. When it is noted that immediately preceding this statement he had said that he could not remember if it was the accused or Baca-flor who opened the door, this point is emphatically established, to my mind. It was not to be reasonably expected that his memory would have improved in the interim and if the expression had become positive, it was most unlikely to be reliable. The appropriate question for the Crown to ask, on such a crucial matter, was whether the witness could now recall how he had got into the building; this was, at all events, the logically preceding matter. Having, decided to seek the unqualified answer, if he could get it, it seems to me that the Crown deliberately invited the cross-examination that occurred. It seems to me to be most unfair that the Crown should seek to take advantage of an unqualified answer that he knew was unlikely to be true upon the basis that the most he could lose was that the cross-examiner might, on the basis of the statement, bring the witness back to a state of uncertainty. In the result, however (and, no doubt on the basis of instructions) the cross-examiner sought to adduce the evidence in a way that was based upon a positive recollection, building upon the supposition that nothing out of the ordinary happened that night.

I think it would be most unfortunate if counsel, especially prosecuting counsel, thought that s38 of the Evidence Act might be used a device to nullify the effect of a cross-examination where they have themselves proceeded or permitted the evidence to proceed upon a false or unreliable basis. I agree that, in the peculiar circumstances of this case, the learned trial judge has not been demonstrated to have erred in the exercise of his discretion in permitting the further examination by the Crown prosecutor although (with unfeigned respect for his Honour) I do not say that I would necessarily have come to the same conclusion.

It should be noted that s38(1)(a) is confined to circumstances in which a witness gives "unfavourable" evidence. Other paragraphs of the subsection give other grounds for cross-examination, of which para (b) is directly applicable to the circumstances here and which would have permitted the Crown to ensure that this important evidence went to the jury upon an accurate basis. However, as I have said, he did not choose this course. It seems to me that some attention will need to be given in due course to the meaning of "unfavourable" so far as the Crown case is concerned. The Crown case is, in essence, the truth, wherever that might lead and even if it leads to a reasonable doubt about guilt. I am far from persuaded that merely because a witness declines to give evidence supporting the theory of the facts for which the prosecution contends or, indeed, gives evidence that contradicts that theory or contention, his or her evidence may thereby be regarded as "unfavourable". How can truthful evidence ever be "unfavourable" from the Crown's point of view? In the circumstances of this case, however, there were good reasons for concluding that, if not untruthful (and I do not think that this was ever suggested in the sense of being deliberate) the evidence of the witness as adduced by the cross-examination was doubtful, even if (by parity of reasoning) the evidence adduced in chief by the Crown was also doubtful.

In Penney v The Queen [1998] HCA 51 (unreported 13 August 1998) it was submitted for the appellant that the trial judge had erred in omitting to instruct the jury that, if motive was relied on by the Crown in proof of the charge (of attempted murder), then it must be proved beyond reasonable doubt. Support for this proposition was gained from R v Murphy (1985) 4 NSWLR 42 at 59-60. Penney was brought to our attention by Buddin SC, for the appellant, after argument had concluded and whilst this Court had reserved its judgment. As is clear from James J's judgment, the same argument as had been put to the High Court was put to this Court. I respectfully agree with what James J has said about it.

Penney does not advance the appellant's case. Callinan J (with whom the other judges agreed) distinguished Murphy upon the basis that the relevant portion of the summing up "was not dealing with intention or motive as such but was pointing to evidence with respect to the relationship between the appellant and his wife upon which [the jury] for an ultimate conclusion of guilt beyond reasonable doubt", citing the well known passage from the judgment of Barwick CJ in Wilson v The Queen [1970] HCA 17; (1970) 123 CLR 334. I do not read the judgment of Callinan J as endorsing the passage from Murphy.

It seems to me that, especially having regard to the explanation of Chamberlain v The Queen (No2) (1984) 153 CLR 520 expressed in Shepherd v The Queen [1990] HCA 56; (1990) 170 CLR 573, no direction along the lines submitted by counsel for the appellant was called for in this case. There may be cases depending upon circumstantial evidence in which proof of motive is an indispensable link in the chain of proof of the guilt of the accused and thus will need to be established beyond reasonable doubt. However, such cases must be rare. This case was certainly not one of them. It is always important to note that the process of reason used to move to a verdict is a matter for the jury. Thus, in some cases it may be necessary to direct them that, if they were of the view that any fact, such as motive, was, so far as they were concerned, an indispensable link in proof of guilt, then they must be satisfied of it beyond reasonable doubt before they could convict. Again, however, I do not think that this was such a case.

Accordingly, I agree with James J that the seventh ground of appeal should be rejected.