Forbes v The Queen
[2010] HCATrans 120
[2010] HCATrans 120
IN THE HIGH COURT OF AUSTRALIA
Office of the Registry
No C10 of 2009
B e t w e e n -
BENJAMIN JAMES FORBES
Applicant
and
THE QUEEN
Respondent
FRENCH CJ
HAYNE J
HEYDON J
KIEFEL J
BELL J
TRANSCRIPT OF PROCEEDINGS
AT CANBERRA ON TUESDAY, 18 MAY 2010, AT 10.16 AM
Copyright in the High Court of Australia
MR P.S. HASTINGS, QC: May it please your Honours, I appear with my learned friend, MR S.L. GILL, and my learned friend, MS K.J. EDWARDS, for the applicant. (instructed by Ken Cush & Associates)
MR B.W. WALKER, SC: May it please the Court, I appear with my learned friend, MR A. DOIG, for the respondent. (instructed by Director of Public Prosecutions (ACT))
FRENCH CJ: Yes, Mr Hastings.
MR HASTINGS: As your Honours would be aware, the matter is an application for leave to appeal against the conviction of the applicant for an offence of sexual assault which was upheld by the Court of Appeal in the Australian Capital Territory. It is the basic contention of the applicant that the conviction was unsafe because the only evidence against him was DNA evidence. We submit that in principle he should not have been convicted on DNA evidence alone, but particularly in the facts of this case where there was additional evidence which was in part exculpatory he should not have been convicted.
The fundamental premise to our submission is that there was no other evidence than the DNA. In our submission, that is apparent from the facts of the case. It is also apparent from the manner in which the trial was conducted by the Crown Prosecutor. In the additional material provided by the respondent there is a copy of the closing address of the learned Crown Prosecutor in which he quite squarely summarised the case at page 183 of the booklet entitled “Respondent’s Materials” as a neat issue between what he described as science and the testimony of the applicant and his partner. At the top of page 183 your Honours will see that the learned Crown Prosecutor said:
At the end of the day, does the science tell you, that the only person who could have committed the act on [Ms K] on 11 March 2005 . . . in that desolate area, was Mr Forbes, or does his explanation that he was at home that evening, win you over.
Again, he goes on to talk, in the next paragraph, about whether they have shown that science was wrong.
In any event, your Honours, we would say that it is clear enough from the facts that there was no other evidence which was capable of incriminating the applicant than the DNA and it does squarely come before this Court as a matter which the Court can consider in principle as to whether the evidence of DNA alone is capable of establishing the guilt of an accused and, in particular, on the facts of this case.
FRENCH CJ: Well, it is the facts of this case we are concerned with and is the evidence that was before the jury when you speak of DNA evidence - I take it that you are using that as a shorthand for the evidence of the taking and matching of samples and the significance thereof given at this trial?
MR HASTINGS: Yes, your Honour. I will take your Honours to the precise evidence shortly.
HAYNE J: The larger proposition you advance assumes necessarily, does it not, that DNA evidence could never rise higher than it did in this trial?
MR HASTINGS: Yes, your Honour.
HAYNE J: Why should we embrace that proposition is I think the question you need to confront when you come to that larger question. But we begin, do we not, with what was the evidence at this trial?
MR HASTINGS: Yes. Indeed, your Honour, but it would seem that the evidence on the topic of DNA was typical, in the sense of outlying principles and theories upon which DNA has been based since it was first acknowledged as evidence capable of being received as expert evidence.
HAYNE J: That makes a number of assumptions about what this jury was told by the witnesses who gave evidence at this trial which, at least for my own part, requires some demonstration because I should say to you at once,
Mr Hastings, that it is not evident to me exactly what it was that the jury was told at this trial beyond the fact that the relevant samples provided a match. They could go to that point, but beyond that it is by no means clear to me what the evidence showed.
MR HASTINGS: I will take your Honours to the evidence in due course. Might I just say at the outset, it is of secondary importance to my client as to whether this represents a matter of broad principle or whether, simply on the facts of this case, the facts attract a general ‑ ‑ ‑
FRENCH CJ: Probably of no importance at all.
MR HASTINGS: That is right. Can I just summarise the way in which we submit the DNA evidence as led leads to the conclusion that we advance. As the evidence will reveal when I go to it, the fundamental premise of DNA evidence as outlined by the expert, Mr Walsh, was that it provides an estimate of the probability that one hypothesis was more likely than another.
The evidence of DNA does not prove facts by itself, it is clearly inherently limited. It is not an expression in itself of real or known facts but rather is a statistical projection based upon data taken from a comparatively small number of people from the population. What we submit that leads to is that in a vacuum DNA evidence does not entitle any conclusion on facts at all, it needs to be assimilated into the evidence of the facts of the trial in order to give the estimate which has been provided of a theoretical nature some factual weight in the context of the issues in the trial.
HEYDON J: So you are premising your present submission on an attack on the low number of the people in the ACT database, 400 odd or 600 odd, whatever it was?
MR HASTINGS: I am not attacking it your Honour. It was accepted by the evidence of the expert that that was a satisfactory number for the database.
HEYDON J: It was an estimate or a hypothesis derived from data based on a comparatively small number of people. What would be a comparatively satisfactory number of people?
MR HASTINGS: Yes, there was no challenge to the adequacy of the database. My point simply is that it is a very small database from which figures are extrapolated in order to give huge figures as to the frequency with which certain profiles appear within the population.
HEYDON J: In the United Kingdom it is 3 million apparently. What would a satisfactory figure be for the ACT, or a better figure than 400 or 600?
MR HASTINGS: It does not matter, your Honour. The point is that it is not a survey of the entire population.
HEYDON J: Anything less than totality is fragile.
MR HASTINGS: I am not sure that I need to say “fragile”. It just puts the evidence into context to note that it is based on a quite small number of people and then becomes a theoretical estimate of how that pattern is to be reflected by the whole population of the community in which the evidence is given.
BELL J: Another way of describing the process that the expert looking at the results of the matching arrives at instead of a likelihood ratio is an opinion concerning the expected frequency of such a genotype in the general population occurring at random. Is that so?
MR HASTINGS: Yes.
BELL J: So that to raise the question that Justice Hayne directed your attention to earlier, in some cases it might be, having regard to the variables in a given genotype, that the expected frequency will be expressed in terms of one in a million, for example, which may be the evidence that was given here, and that might lead to a certain conclusion concerning the ability of evidence of that character alone to sustain a conviction that would be different to a genotype of greater rarity, as is commonly found in evidence of this sort. It just seems to me to create a, perhaps, difficulty with your broader proposition.
MR HASTINGS: With resect, your Honour, we would see the issue the same and that is that it is just a matter of theoretical estimate as to the frequency with which the profile occurs in the population because the only information which is known comes from the 620‑odd entries in the database. To anticipate the frequency across the population at large is a matter of theoretical and statistical calculation rather than dealing with a real fact.
BELL J: Appreciating that, there is a material difference in an expected frequency of one in a billion persons and one in 10,000 persons. Why cannot a jury evaluate those propositions for what they are worth?
MR HASTINGS: They can, your Honour, but we submit based upon a tangible fact in the case which links the applicant to the situation in the first place. This is not squarely answering your Honour’s concern, I know, but can I just finish and come to what we see to be the real crux of the legal issue and that goes back to the fundamental task of the DNA expert in providing a figure which provides the estimate of the probability that one hypothesis is more likely than the other. What we submit in the context of the criminal trial is that one cannot do that in the abstract or in a vacuum, because to enable the Crown to have its hypothesis assessed and evaluated in this manner means that there must be some basis for the Crown to put forward that hypothesis.
HAYNE J: There are several steps in that which I think may need to be teased out in quite some detail, Mr Hastings. First, that form of analysis differs, does it not, from a frequency analysis? Frequency analysis and likelihood ratios are two distinct forms of analysis, are they not?
MR HASTINGS: Yes.
HAYNE J: So you have gone now to a likelihood ratio analysis, have you?
MR HASTINGS: That is how I understand the evidence of the expert to be expressed, your Honour.
HAYNE J: Yes, and the meaning to be attributed to that evidence depends, does it not, in part, upon what is meant – what are the hypotheses that are being compared. Is that right?
MR HASTINGS: Yes.
HAYNE J: Were those hypotheses ever identified with any particularity at this trial?
MR HASTINGS: They were clear enough, your Honour. One was that the accused was the person who was responsible for leaving the samples on the clothing of the victim. The other hypothesis was that according to the evidence of the accused, that it was not him.
HAYNE J: But are those the hypotheses which I think Mr Walsh was the witness concerned, perhaps Ms Pearson, but are those the hypotheses those witnesses were comparing?
MR HASTINGS: Yes, your Honour. I will come to Mr Walsh’s evidence on the topic, but can I just not be deflected yet from my basic proposition which is a quite basic forensic assertion and that is that in the criminal trial context, the Crown should not be allowed to put forward a hypothesis without a factual basis for it, which in turn of course leads back to our basic proposition and that is that DNA alone should not be a basis for conviction. There should be some other evidence which would justify the finding of guilt against the accused, not on its own but certainly complement the DNA evidence.
I draw a parallel in a very general way with the way in which the Court has laid down a rule about circumstantial evidence in the sense that it is clearly to be understood that a person is not to be convicted on the basis of circumstantial evidence unless it is capable of excluding a reasonable hypothesis of innocence.
Now, that is a clumsy parallel but, nevertheless, it is a concept which we would seek to import into this situation by asserting that the Crown should not be allowed to put forward a hypothesis to attract the application of the DNA evidence unless there is a factual and evidentiary basis for it, otherwise it becomes a complete bootstrap argument where the Crown, with no evidence at all, says “We put forward our hypothesis on the basis of DNA evidence and now we will rely upon the DNA evidence to give a likelihood of how correct that is”. In my submission, it leads to a recognition of the fact that the Crown should not be entitled to put forward a hypothesis of guilt unless there is some factual basis for it other than the evidence that it then seeks to invoke.
HEYDON J: Does that not mean that the DNA evidence was inadmissible?
MR HASTINGS: It could have been, your Honour; yes. There were various points at which this point could have been taken. One of them was at the end of the evidence – just to take a voir dire point I suppose and say this evidence is never capable of establishing the guilt of the accused. The other would be to take a no‑case point at the end of the evidence.
HEYDON J: There would be some practical difficulties in some cases because you could never be sure whether or not there was other evidence.
MR HASTINGS: That is right. You could raise the point ‑ ‑ ‑
HEYDON J: It would depend on what the jury might make of it – let us put on one side that possibility with this case but apart from disbelief of the accused, I do not believe any other evidence has been pointed to. It could not have been pointed to from the close of the Crown case.
MR HASTINGS: No. Your Honour, I am quick to concede that there were various alternative avenues available to the accused in this matter which could have led to the same result, but in the end it was allowed to go to the jury. He was convicted and the point was taken on appeal. It is from that judgment that we seek leave to appeal.
BELL J: Coming back to your broader proposition, Mr Hastings, if one were to look at other areas of expert evidence, and for present purposes at fingerprinting, underlying the expression of opinion that one fingerprint is identical to that of the accused is the assumption that no two people in the world have the same fingerprint. I am not aware that that has ever been established as a fact.
The evidence concerning DNA is commonly given not in terms of an expression of opinion of identity for the reason both that the whole genome is not tested and the entire population of the world has not been tested. So as a matter of rigorous science such an opinion cannot be expressed. But it is still not clear to me why, assuming the evidence is able to be tested and the results obtained by the Crown independently verified by testing by a defence expert, if the frequency in the expected population is of a sufficient rarity it is not open to conclude on that alone, beyond reasonable doubt, that the donor of the sample is the accused.
MR HASTINGS: Your Honour has raised the question of fingerprints and it is true, obviously, that there is not an exhaustive analysis of all the fingerprints in the world to substantiate what seems, however, to be an accepted position that fingerprints are unique. So there is that underlying premise which justifies fingerprint evidence but that is not the case in relation to DNA because there is a recurring frequency of DNA profiles within the limited survey that is conducted of the human DNA strand.
BELL J: The matter that I am taking up with you really reflects the more rigorous approach as far as the science in DNA analysis is concerned, does it not? One comes back to the point that with the fingerprint, one assumes that is correct to say that if you have 12 or nine or however many points of identity, you will not find another human being presenting such a profile.
MR HASTINGS: Yes.
BELL J: When you speak of this evidence as being in some way lacking a real quality you do not seem to come to terms with the fact that the DNA evidence itself will vary in terms of its cogency and that will depend in instances on the relative rarity of the combination of nine loci and the degree of variation that is present in the expected population.
MR HASTINGS: Well, I am sorry, your Honour, I certainly accept that proposition and that is that the statistics can vary from modest return to what can be mind numbing almost in its odds, I suppose. I certainly accept that but in our submission, your Honour, it still does not change the fundamental character of the evidence that it, by giving that figure, has ventured into a realm of estimate rather than dealing with something which is factual.
FRENCH CJ: Just going back to the focus on the evidence that was given to the jury and, I think, we need to look at precisely what they were told, in relation to Ms Pearson’s evidence, I do not think she actually mentioned any numbers, did she? She just gave the verbal scales.
MR HASTINGS: Yes.
FRENCH CJ: So that what the jury were told in the course of her evidence was that results indicated an either strong, in relation to one of the samples, or extremely strong, in relation to the other sample, probability that they came from some other source. Was that anything more than a kind of taxonomic label on the numbers? I do not think she was expressing it as an independent opinion. Is that right?
MR HASTINGS: It was an accepted formula by reference to scales of figures.
FRENCH CJ: So if it is more than one in a million it is called extremely strong, if it is from 100,000 up or whatever, it is just strong.
MR HASTINGS: It was very strong and there was a lower limit of over 10,000 which was strong. Well, your Honour, that was an accepted course which was adopted in the trial.
FRENCH CJ: Yes. She had gone through the application of the appropriate equations, the appropriate multiplications, as it were, and she had come up with a figure. This was not given to the jury by her but rather the label attached to that figure.
MR HASTINGS: That is right.
FRENCH CJ: What more did Mr Walsh then tell the jury?
MR HASTINGS: He told the jury more in terms of the premises upon which the calculations had been based. It may be convenient if I can take your Honours to Mr Walsh’s evidence on the general principles now and, in particular, picking up the point your Honour the Chief Justice has just raised at page 146 of the appeal book at around line 30.
HAYNE J: The numbering of this appeal book is deeply confusing. I think the real numbering is the typed numbering at the foot, Mr Hastings, is it not?
MR HASTINGS: I am sorry, your Honour, yes, it is.
HAYNE J: We have about 27 numbers on every page.
MR HASTINGS: I am sorry, your Honour, yes. Referring then to page 133 of the appeal book, at line 30 Mr Walsh dealt with the matters that the Chief Justice was just referring to, and the answer which follows at line 32, in our submission, is significant and in a way captures the crux of his evidence about the fundamental principles ‑ ‑ ‑
FRENCH CJ: Before you go to that line, he was not actually telling the jury what the numbers were in this case, but rather the significance of the verbal scale.
MR HASTINGS: Except that at the same time it was a lead in to what the evidence meant in any event because he went on to say:
The likelihood ratio –
without necessarily dealing with the terminology which was being used –
is the name for the statistic that we calculate. It’s called a ratio because it compares the outcome of two different probability estimates, and it weighs one against the other and creates a ratio. The probability estimates are based around some hypotheses that in the legal context we typically call the prosecution hypothesis and the defence hypothesis.
Really, your Honour, that is the point to which my submission earlier was directed and that is that in order to attract that comparison, the prosecution hypothesis needs to have a justification in the form of an evidentiary basis which recalls some evidence other than the DNA itself.
HAYNE J: Understanding that to be a point that you make, can I look at the answer at 133 and ask you whether elsewhere in the evidence of Mr Walsh or Ms Pearson led before the jury content is given to prosecution hypothesis and defence hypothesis. Can I also ask you whether in the evidence of either of those witnesses, content is given to the expression he chose “based around some hypotheses”, that is, my base question is what where the jury going to make of what they were told by Mr Walsh in this answer?
MR HASTINGS: I do not think there is a short answer to what your Honour has asked me other than it was fundamental to the conduct of the trial that there were two cases, the prosecution case and the defence case. The prosecution case was that it was the accused who committed the offence because his DNA was matched with the samples taken from the victim, whereas the defence case was that it was not me, it must have been somebody else.
HAYNE J: I understand that. I understand that was the forensic contest, but what were the jury to make of this answer. They are told that they are dealing with a likelihood ratio. They are told that that is a ratio of two different probability estimates. They are told that the estimates are “based around” – I do not know what a jury was going to make of that – “are based around some hypotheses”.
Are they told at any point in the evidence what those hypotheses are, that is, is it explained to the jury what the steps in reasoning are that they are being invited to take? If they are not, then that is one set of issues we have to confront. If they are told what steps they are being invited, by the prosecution, to take from the premise that we can observe identical DNA profiles to the conclusion asserted by the prosecution, “From observation of identical profiles you can conclude that the accused left the biological material that generated the profile from the garments”, then I can begin to understand it. But unless we take those steps, Mr Hastings, I am left simply with a collection of words.
MR HASTINGS: In a sense, I suppose, your Honour, it is in the applicant’s favour if these matters are not explained to the jury because it enhances the risk that the jury does not understand this evidence, in the first place, if they are not given the assistance of ‑ ‑ ‑
HAYNE J: It is not a case of understanding, it is a case of what was the evidence that was before the jury which they were invited to use to reason from a premise, a premise, namely that the two profiles are identical, to a conclusion not only are the profiles identical, but one of them recovered from the complainant’s garments was left there by the accused.
MR HASTINGS: I am reminded, your Honour, at page 114 in the course of Ms Pearson’s evidence, who gave a more fulsome answer than had been given by Mr Walsh, at line 24 in cross-examination she was asked a question which led to the answer:
Well, that’s why we do a statistical calculation. And in the statistical calculation we look at two different hypotheses, so the first one may be that the person contributed to the mixed DNA profile, and the second hypothesis might be that another person - or two other people chosen at random from the general ACT population may have contributed to that mixed DNA profile.
BELL J: But was that an explanation with respect to the sample that was a mixed sample?
MR HASTINGS: It was in that context, your Honour; yes.
HAYNE J: All cast in terms of may be and might be. The hypothesis may be this, might be that.
MR HASTINGS: Your Honour, at page 272 in the course of the summing‑up the trial judge referred to the evidence of Ms Pearson. At line 20, the paragraph begins with a general reference to DNA testing and the techniques. In relation to the sample which is 2A@, which was the sample from the black pants, his Honour reminds them of a caveat which:
Provided extremely strong evidence to support the contention that the DNA reference sample relating to the accused was the source of the –
DNA obtained from that sample. I am not sure that I can point your Honour to any more ‑ ‑ ‑
BELL J: I think at 157, going over to 158, Mr Walsh explains in general terms the approach to the assessment of the frequency of the appearance of such a genotype. Then he explains the reasons why one then multiplies the figures for the expected – arrived at at each allele to obtain the ultimate figure, which I think there is a figure as to the expected frequency within the population of that genotype.
MR HASTINGS: Yes, that is so, your Honour. In answer to your Honour Justice Hayne, I am not able to find, even in the summing‑up, any more direct references to what it was that Mr Walsh was talking about. However, might I nevertheless take your Honours to some further statements of general principle by Mr Walsh. That includes over the page, at page 134, from where I was previously when he was asked about what “strong” and “extremely strong” meant and at line 6 he was asked:
Now these are statistical figures, are they not?‑‑‑Yes.
What does it mean when you say greater than 10,000, or greater than 1 million, from a statistics point of view?—Well, that’s – I guess from a statistics point of view it’s simply the outcome of the statistical – particular statistical evaluation we’ve undertaken . . .
No, it doesn’t. That figure, if we were to use a probability as you have there of 1 in a million, it can be used to provide an estimate of how often you might be expected to observe that particular DNA profile as we talk about in this circumstance. But it doesn’t allow you to make a prediction, or to actually say exactly how many you would see in a population of for example 1 million.
At page 137 at line 16 he again gave a further general answer. Without quoting the question, he said:
No. That’s correct. We’re able to use those probabilities as I said as a way of estimating, if you like the commonality or in many cases, the rarity of a particular profile in a particular population. But they can’t be used to provide an exact number. They are estimates. They’re statistical estimates only. And they aren’t real numbers, they aren’t real observations. So it’s important not to translate them from the estimates that they are to actual real expectations or real observed numbers.
He repeats the same point in the next paragraph at around line 34 when he says:
You have to be cautious to remember that the statistical probability is a statistical probability and it is not an observed exact number.
FRENCH CJ: Incidentally, so far as the jury was concerned, the “extremely strong” designation given to one set of results by Ms Pearson, as explained by Mr Walsh, could have implied a figure of one in a million or one in a billion.
MR HASTINGS: Yes, it was in excess of the minimum.
FRENCH CJ: But they did not know what the figure was.
MR HASTINGS: No. I will not weary your Honours with too much of Mr Walsh’s sometimes lengthy answers, although in some respects they were useful in terms of, at least by repetition, giving the jury a fuller opportunity to understand the concepts being spoken of. At page 142 he was taken back to the significance of “extremely strong” and again, if I can just read part of the answer, at line 12:
you mentioned at the early part of this discussion the first assessment that’s made by a scientist in this area is to look to see if an individual can be excluded on the basis of their profile. Because that can be an absolute decision, a decision whereby that person is no longer considered as associated with that item of evidence. When you have a situation where a person on basis of their profile cannot be excluded, because their profile corresponds with the profile observed in a crime sample, and then as I mentioned on Tuesday that’s when we need to talk in statistical terms about the strength of that particular association. One thing that we cannot say as scientists, with certainty, is that what we have observed is absolutely what has happened. We talk in terms of the DNA results because those are the results that we’ve produced, and that is in essence our evidence. It’s not always possible, in many cases it’s not possible, for us to extrapolate our conclusions from that DNA environment to an environment where we talk about perhaps what might have been alleged to have occurred at the offence level or in the interaction between the complainant and the perpetrator. So we’re very careful what we say to retain our comments or restrict our comments so that they deal with the DNA results that we’ve obtained and don’t sort of trespass, if you like, into the area that is related to the actions or activities that might have occurred at the offence – at the time of the offence or things of that nature.
Then the next answer which commences at line 38:
Yes, that’s also another component to that, and another reason why as scientists we don’t speak in absolute terms, because our results that we present are based on a limited set of [facts], in this case 10 tests, and we haven’t tested the DNA exhaustively, and we can’t say therefore that the entire DNA, including those areas we haven’t tested, would also match. So we restrict ourselves again on the basis of the tests that we have undertaken.
In a way, I suppose, that captures what has already been accepted and that is that there is this significant restriction or limitation on the evidence itself which, in our submission, means that unless there is a factual base to which the formula is to be applied, then it cannot have a meaning in a particular case.
FRENCH CJ: At page 192, Mr Walsh in cross examination gave an answer at about line 10 to the question or the proposition if all the alleles all match up, it does not mean that it is that person’s DNA. Then he said:
No, we can’t say, as we discussed earlier . . . that that person is categorically the source of that DNA, just that on the basis of those results, they – they can’t be excluded as being a possible source.
Then the next question, as it were, restated that proposition and he answered again in the affirmative. Is that consistent with the evidence that he had given previously or was the evidence that he had given previously stronger than that in terms of the use to which this evidence could be put?
MR HASTINGS: Well, your Honour, I thought it was consistent with the paragraph that I recently read from page 142 where he dealt with the position which you can absolutely exclude somebody where there is not a match.
FRENCH CJ: You can exclude somebody who is not a match. What he is saying here, it seemed to me to be, that if there is a match, the most you can say is that person is not excluded. Is that what he is saying or am I misunderstanding it?
MR HASTINGS: No, I think that was what he was saying, your Honour, but then, of course, it becomes a matter of statistical estimate as to the extent to which he may be included.
BELL J: And absent taking or mapping the genome of the entire population of the world, as I understand Mr Walsh, that would always be the case, that is, that one could say no more than that an individual was not excluded no matter how rare the particular genotype was.
MR HASTINGS: With respect, your Honour, that may be overstating the situation. I mean, one could increase the number of loci, for example, presumably, and get closer to a point at which one could be certain. I mean, I am drawing again on the fingerprint precedent that experts accept, it seems, without question that fingerprints are unique. I mean, I am not the person to judge, but it would seem to me that without surveying the world one could just increase the number of loci examined to virtually reach the same point but the way the program works it selects nine plus the gender loci in order to carry out the calculations.
BELL J: Yes. I think somewhere there was reference, either in the evidence of Ms Pearson or Mr Walsh, to the difficulty created by not having mapped the DNA of every person in the world.
MR HASTINGS: I think that was Mr Walsh, your Honour. Yes, that is right. Indeed, Ms Pearson may have said the same thing.
HAYNE J: But the debate we have just been having demonstrates does it not, that the use of DNA evidence by a jury will depend upon a series of steps. In this case we focus upon two. The two steps in issue in this case are, first, are the profiles obtained from the reference sample and obtained from the complainant’s clothing identical? Step one. By the time the case went to the jury it was the accused’s speech, as I understood it, that there was no dispute that the profiles were identical. But the second step that has to be taken by a jury is going from the observation of identity of profile to a conclusion about whether it was the accused who deposited the biological material recovered from the complainant’s clothing, and it is in connection with that second step that we have evidence given about statistics. Is that right?
MR HASTINGS: Yes.
HAYNE J: That statistical evidence seems to have, in part, made reference to questions of frequency, in terms of one in a million, one in a whatever number was inserted, but also seems to have invoked a different notion – a notion of likelihood ratios. Is that right?
MR HASTINGS: Yes.
HAYNE J: As I understand it, the Crown went to the jury on the footing of likelihood ratios.
MR HASTINGS: Yes.
HAYNE J: The evidence that you have thus far taken us to about likelihood ratios and what they mean includes, perhaps centres upon the answer given by Mr Walsh at page 133.
MR HASTINGS: Yes.
HAYNE J: A question which emerges is whether that was enough to put before a jury to permit a conclusion of persuasion beyond reasonable doubt.
MR HASTINGS: If the answer to that is no it was not enough, then I embrace it readily, but I am not sure ‑ ‑ ‑
HAYNE J: You astonish me, Mr Hastings.
MR HASTINGS: I am not sure that I can endorse it in a positive sense for the reason that I simply gave earlier, your Honour, that I just think it was clearly understood throughout the trial that there were two versions of events – one which supported the accused being involved and the other that he was not. That issue was the issue in the trial.
HEYDON J: Your case is really that that which supported the Crown’s versions of events is a nothing. It is just an abstraction, it is mad scientist stuff. If that is so, why does it make a difference that there might be some evidence against the accused? I may have misapprehended the way you have been putting your argument and perhaps the way similar arguments have been put in the cases in the past, but one view of it is that the proposition is you cannot rely on this evidence unless it is corroborated or you cannot rely on this evidence unless there is independent evidence of guilt. It is a very rare proposition, probably non‑existent in our law outside statute that evidence is admissible if corroborated but not if it is not. Why does the existence of some other evidence, for example, if your client had been seen nearby 10 minutes earlier apparently innocently engaged, why does that suddenly make all this other evidence fit to go to a jury and be the basis of a conclusion of guilt?
MR HASTINGS: Your Honour, my non‑technical answer and legalistic answer, I think, is that the Crown need a factual basis for advancing their hypothesis and the Crown should not be allowed to advance a hypothesis without some justification and then rely on DNA to justify it. It is a somewhat simplistic proposition, your Honour, but nevertheless, in terms of a criminal trial, in my submission, the Crown should not just be allowed to pluck a hypothesis out of the air and put it to the expert and say, well, what do you say about the involvement of the accused now, without there being a proper basis in the evidence for that hypothesis to be available to be considered.
HEYDON J: As you said earlier, a factual base to which the formula can be applied. Is that really how it works?
MR HASTINGS: Your Honour, in my, again, somewhat crude analogy, it would seem to be it is like using a multiplier to nought. If you multiply nought by nothing you get nothing. However, if you give the subject a value even of one, all of a sudden the statistical figure when multiplied gives a result, but unless you have something which justifies the application of the formula in the first place, then one should not be entitled to get any more than zero, which is where one starts.
FRENCH CJ: Why cannot the Crown say, “We took a sample from the complainant’s clothing. We took a sample from your client. We found a match which occurs randomly at a certain frequency. A reasonable hypothesis to be constructed upon that – that is supportive of the hypothesis that your client was present at the time of the offence.”? The question that then has to be addressed is whether there is another reasonable hypothesis consistent with innocence, even if disbelieved. That is the standard approach to this sort of thing, is it not? We are not constructing a hypothesis first and using the DNA to justify it. They are using the DNA to support a hypothesis. That is the way they are approaching it.
MR HASTINGS: Well, your Honour, it would seem to me that they are using the DNA to give them a hypothesis in the first place and then using the DNA to bolster it by saying, well, now we ‑ ‑ ‑
KIEFEL J: What is the hypothesis that you are referring to in this regard?
MR HASTINGS: That it is the accused who is the offender.
KIEFEL J: Rather than the hypothesis being that he is the person who has deposited the material which has been analysed?
MR HASTINGS: I am sorry, your Honour. Yes, your Honour is correct. I mean, technically that it the first step, but the second step follows almost automatically, I would think, in a case such ‑ ‑ ‑
HAYNE J: Not in every case.
MR HASTINGS: Not in this case, your Honour.
KIEFEL J: But if the hypothesis is that he is the person that has deposited the material, why is a factual premise such as, he was seen in the vicinity, any necessary premises for the hypothesis?
MR HASTINGS: Because it is a starting point to having some involvement between the accused and the offence, otherwise there is none, apart from the fact that a DNA analysis was carried out which purports to identify him with a matching profile and that therefore there is a million to one chance that it was ‑ ‑ ‑
KIEFEL J: But the first is a circumstance. The factual premise may be a circumstance. The hypothesis is not simply a circumstance, though, is it? It is more than that.
MR HASTINGS: The hypothesis, as I understand it, is to be the Crown theory to the effect that the accused is the offender.
BELL J: Mr Hastings, if, instead of going for the larger proposition, one were to look at the smaller proposition, which is that this verdict was unsafe, Mr Crown went to the jury on the basis of asking, does the science tell you that the only person who could have committed the act was the accused, the evidence in the way that it was presented, and looking at this for present purposes on a view that at one stage Mr Walsh’s evidence was consistent with a view of an expected random match of one in a million, then absent any other circumstance, is not your point that that lacked sufficient cogency to support a verdict beyond reasonable doubt since, whilst it showed the accused was likely to be the contributor of the sample equally in a population the size of Australia, it might be assumed that there were others who would answer the description of being a sexually active male who, on the statistical evidence that was presented, one might expect to find had the same genotype?
MR HASTINGS: Yes. We certainly, your Honour, have a second limb to our argument away from the general principle and that is that on the facts of this case it was unsafe because there were other factors which did not include, and whether they were exculpatory or not is a matter of debate, but they certainly did not include the applicant as the person who had committed the offence and by reference to the descriptions given by the victim and so forth.
BELL J: The matter that I am raising with you is simply that the evidence, insofar as it went of relative rarity, fell short of enabling a jury to conclude identity beyond reasonable doubt, absent any other circumstance. Although it is to be noted that one circumstance was that he contributed the DNA sample on an occasion when he was present in Lyneham and Lyneham was the scene of the offence. So I suppose that might figure in a jury’s approach to the evidence.
MR HASTINGS: Yes. Although I did not see any great reliance upon that circumstance in the way the case was conducted.
BELL J: No.
MR HASTINGS: I otherwise accept what your Honour has indicated. Your Honours have no doubt noted the terms of Ms Pearson’s evidence in relation to the particular items. I am not sure that I would gain anything by going to her evidence. She used the same terminology, of course, and in simpler terms used the same explanations as were given with perhaps more authority by Mr Walsh. As your Honours have observed, the way the matter was assessed was on the basis of descriptions, strong, very strong or extremely strong, rather than by statistical information. That was done by acquiescence, certainly on the part of the defence, and is not something about which we can complain. It does, however, lead to an interesting outcome in that it has the witness then expressing the opinion that it was extremely strong evidence that the sample had come from the applicant.
FRENCH CJ: Is that what the witness did? Even in the terms in which the evidence was given, the witness did nothing more, did she, than to apply the verbal scale? As I understand it, she was not expressing some independent opinion or judgment about the significance of the figures. She was saying this is how we have labelled figures in this range or not “we”, but this is how they have been labelled by relevant experts.
MR HASTINGS: Your Honour, I had in mind, in particular, the answer at page 89, line 35 when dealing with her conclusions about the result of the comparison of the Forbes reference sample and the material from the pants, which was sample 2A, at line 35 she said:
In my opinion, these findings, when considered in isolation from other information, provide extremely strong evidence to support the contention that the donor of the DNA reference sample relating to Benjamin James Forbes is the source of the DNA profile obtained from ‑ ‑ ‑
FRENCH CJ: But that conclusion was simply an application of the verbal scale, was it not?
MR HASTINGS: Yes. It leads to, in a way – it is stronger than curious, but the alternative statistical expression is that it is a million to one more likely to be one person than a person at random.
FRENCH CJ: But what she did was she crunched the numbers applying a formula which had been developed by others and got the figures and, as I understand it, applied the relevant verbal scale.
MR HASTINGS: Yes.
FRENCH CJ: What appears in the report is not an expression of opinion about the strength of the inferences to be drawn from those figures. It is just the application, as Mr Walsh later explains, and I think she explained also in her evidence, of that verbal scale.
MR HASTINGS: Except the terminology used is that it provides extremely strong evidence.
FRENCH CJ: I know that is the terminology, but it was because there was a so‑called verbal equivalent to be delivered to a jury instead of numbers.
MR HASTINGS: The traditional expression is perhaps slightly more sterile in that it would otherwise have been said that it was a million to one more likely that the donor of the sample was from Forbes than from a person at random in the community.
FRENCH CJ: No, all that this tells you is that it is a million to one or more, is it not?
MR HASTINGS: Or more, yes.
FRENCH CJ: Up to a billion to one. It could have been a billion to one.
MR HASTINGS: It could be, yes. I accept that, your Honour, but nevertheless ‑ ‑ ‑
FRENCH CJ: All this under the rubric of “extremely strong”.
MR HASTINGS: Certainly it just seems to be edging closer to the prosecutor’s fallacy that it was the accused who contributed the sample by expressing it as “extremely strong evidence to support the contention that”. As I was saying, your Honours, the alternative statistical expression seems to be a slightly more sterile way of expressing the conclusion than intruding into the case to use terminology which classifies the evidence in the case as being extremely strong.
Your Honours, in our submissions we have set out a series of cases which contained expressions of conditions, in our submission, about the admissibility or the effect of DNA evidence from the time that it was acknowledged as expert evidence in the early 1990s and ‑ ‑ ‑
HEYDON J: You seem to have a decision in your favour of the New South Wales Court of Criminal Appeal, Chief Justice Gleeson, Justice Cripps and Justice Abadee - Green’s Case. That statement by Justice Cripps is the ratio of the case.
MR HASTINGS: Yes, and, your Honour, that was adopted and applied subsequently as well in Pantoja and also in Milat in the passages which we have set out in our submissions. There did seem to be acceptance at that early stage that there was a clearly acknowledged limit on the effect of DNA evidence which would preclude a conviction on DNA evidence alone. The authorities to which we have referred include the Victorian case of Noll which followed shortly thereafter, in which a similar expression was given.
HAYNE J: Does this reliance on these cases depend upon a premise that the science has not altered in the intervening period, and by “the science” I mean the statistical science?
MR HASTINGS: Yes.
HAYNE J: It may be right, it may be wrong. How are we to know whether it is right or wrong?
MR HASTINGS: Your Honours do not. But putting it from our point of view, there is no evidence that technology has changed in any significant way which would justify departure from those statements of principle.
HAYNE J: Do not misunderstand me, I am not referring to the biology of it or the biological analysis generating DNA profiles. What I am referring to is statistical analysis, population, genetics and the like.
MR HASTINGS: There is no evidence as such. In the course of Mr Walsh's evidence, I think he acknowledged that in recent times the database in the ACT had increased from 400 to 620 numerically but there was no indication whether in qualitative terms there had been any change in the way in which the data was collected.
HAYNE J: I understand that, but what I am trying to grapple with, Mr Hastings, from a very imperfect base - the fault is mine not yours - underpinning this evidence I think Mr Walsh said, did he not, that there is some, what was it, heavy-duty science or something, was there?
FRENCH CJ: Heavy theory, I think.
HAYNE J: “Heavy theory”, I think, was his expression. I took that reference to be to statistical analysis, not to the biology of it or to the involvement of PCR analyses or anything of that kind but the statistical analysis is quite complex. Now, you invite us to take account of decisions of courts which, did they turn upon the particular scientific evidence that was led in those cases because if they did, what is the legal principle that we would then have to adopt? If science has moved on in the intervening time then harking back to what was decided by courts on other scientific bases at an earlier time may have some difficulties. I do not say it does but it may.
MR HASTINGS: Well, no doubt technology has improved and the result may be on one view more informative and indeed, more reliable, but the basic tenets of 10 loci and so forth from specified areas within the DNA have been constant, as I understand it, from the time when the authorities earlier expressed their reservations about the capacity of the evidence.
KIEFEL J: My reading of Green’s Case, to which reference has just been made, is that the Crown case stopped at what their Honours there call the eighth point which is the matching point and that the ninth step which is referred to at page 9 which concerned the statistical probabilities, that although the results matched might not be from the same person was not gone into and was not put before the jury and without that step, as I understand the judgment there, a conviction could not stand.
MR HASTINGS: That is so, your Honour, and the respondent has drawn attention to the special circumstances in which that statement was made in Green’s Case. However, I think by the time one moves on to the subsequent cases of Pantoja and Milat that limitation is not evident from those judgments and it seems to be a more broadly applied principle without the limitations which emerge from the facts of that case and similarly when one goes to Noll, the Victorian Supreme Court of Appeal case which we have also referred to, again the case does not seem to be limited by any particular circumstances of the evidence in that case.
FRENCH CJ: That is R v Noll (1999) 3 VR 704.
MR HASTINGS: That is right. The judgment of paragraph 25 cites the New South Wales cases and expresses the general proposition that DNA profiling establishes no more than that the accused could be the offender, not that he or she is the offender and refers then to Doheny and Adams, the English case. The South Australian Court of Criminal Appeal looked at the matter again in 2002 in the decision of Karger, to which we refer in our submissions.
FRENCH CJ: That was Justice Mullighan’s fairly extensive discussion, was it not?
MR HASTINGS: Yes, that is right, your Honour.
BELL J: That was the case where there was the challenge to the Profiler Plus technique.
MR HASTINGS: Yes. Then again caution was expressed about the use of the DNA evidence and the need to examine it in the light of other evidence in the case. We have referred to two subsequent decisions from South Australia of Rowe and Gum as being the two decisions that we were able to find in which the issue had been squarely put at an appellate level that DNA alone was not sufficient, and the decisions both upheld convictions where there was DNA only.
The case of Rowe, I think, caused some confusion in the citation that was used earlier, but it is [2004] SASC 427, and the relevant passage that we have set out in submissions, and we need not repeat, is at paragraph 40 in which the judgment simply concludes without reference to any of the authorities to which I have referred earlier, that it was a safe and satisfactory verdict even though DNA was the only evidence of identification of the appellant.
Similarly, the later decision of Gum, again from South Australia in the Court of Criminal Appeal, a similar outcome was reached, and in paragraph 32 the judgment of Justice Vanstone stated:
Plainly the evidence that DNA matching the appellant’s was found at each “scene” was extremely potent. The jury would have been entitled to view this evidence, standing alone, as sufficient proof of either count.
However, your Honour, we also say about that that there was no apparent reference to earlier authority before that general statement was made and ‑ ‑ ‑
FRENCH CJ: The debate in that case was really about the linkage between the analysis of the reference sample and the profile. There was a linkage question there, was there not, rather than a broader issue?
MR HASTINGS: Yes. We draw them to the Court’s attention because of the general statement which seems to embrace the proposition that DNA alone can be sufficient. We have also referred your Honours to the English cases of Adams and Doheny and Adams which were at about the same time as the earlier New South Wales case. Adams [1996] 2 Cr App R 467, the earlier, did hold that there was no general rule. On page 470 there is a long paragraph which ends with a sentence saying:
There is, however, nothing inherent in the nature of DNA ‑ ‑ ‑
FRENCH CJ: I am sorry, can you just give us the reference you are taking us to there, please?
MR HASTINGS: [1996] 2 Cr App R 467. I am sorry, your Honour, it may be page 469 – I have an electronic copy which was not ‑ ‑ ‑
HEYDON J: It is 467.
MR HASTINGS: The passage at, I think, 469, your Honour. I am sorry, your Honour, it is the paragraph which commences “DNA generally” – no I am sorry:
Mr Thwaites drew our attention to a number of authorities including Re J‑S –
and some others, and then it is a long paragraph, which ends with the statement:
There is, however, nothing inherent in the nature of DNA evidence which makes it inadmissible in itself or which justifies a special, unique rule, that evidence falling into such a category cannot found a conviction in the absence of other evidence.
However, in the decision which followed only a month or so later in Doheny and Adams [1997] 1 Cr App R 369, there is a paragraph which we did not actually quote in our submissions. At page 373, which does, however, contain the same caution which we advocate here, the last paragraph on page 373, around F begins:
The reality is that, provided there is no reason to doubt either the matching data or the statistical conclusion based upon it, the random occurrence ratio deduced from the DNA evidence, when combined with sufficient additional evidence to give it significance ‑
I emphasise those words –
is highly probative. As the art of analysis progresses, it is likely to become more so, and the stage may be reached when a match will be so comprehensive that it will be possible to construct a DNA profile that is unique and which proves the guilt of the defendant without any other evidence. So far as we are aware that stage has not yet been reached.
So that it does contain the qualification which we continue to advocate. Your Honours will have noted that in our submissions in reply we have addressed the same topic again in the light of a more recent statement by Chief Justice Spigelman that some of the reasonable caution may have abated with the improvement in technology. What we have sought to say in a way which is, I have to concede, not strictly relevant to the facts of this case because no issue was taken about contamination, but there has been, as we have endeavoured to show by reference to a number of sources, an increasing awareness of the risk of contamination and as a matter of general principle we would draw attention to those circumstances in order to ‑ ‑ ‑
FRENCH CJ: But that was not an issue in this case, was it?
MR HASTINGS: No, it was not and I am quick to concede that, your Honour, but to the extent that one might otherwise be looking at these issues in the broad and being inclined to assume that as time has passed there has been an improvement in technology which might engender more confidence in the results of DNA testing, that has to be offset by an increasing awareness.
FRENCH CJ: But it is not the technology that is really in issue here either, is it? It is the statistical theory that underlies it and what significance can be attached to the results derived from the application of the statistical theory?
MR HASTINGS: Yes, I agree.
FRENCH CJ: We are working here really on the assumption that everything was done, in terms of the actual physical steps taken, according to Hoyle.
MR HASTINGS: Yes, that is so. To the extent that there may be a general principle under consideration by the court that as a matter of a broad approach DNA alone is sufficient, in our submission, that needs to be viewed with some caution because of the increasing public awareness that there are other risks in DNA evidence than the statistical outcome which is engendered. Your Honour, we have also referred to some other collateral matters, such as the Victorian bench book, indicating that in Victoria, at least, the standard direction to juries is that they should not convict on the basis of DNA evidence alone.
FRENCH CJ: How does that factor into our calculations?
MR HASTINGS: It was a collateral fact relating to the general practice which your Honours may be interested in. Certainly when it comes to deciding this case, we do not suggest that it has any role to play, but to the extent that your Honours might be looking at this at a broader level, we simply bring that to your attention as what seems to be a fact in the real world that there and apparently in the United Kingdom it is the case that prosecutions on DNA alone are not generally run and if they are, they are met with the direction which is to be seen in the case that we cited of Reed
and Garmson where the last paragraph which I referred to refers to the judge’s direction in that case to the effect that the jury should not convict on DNA alone is now reported. We were reminded in [2010] 1 Cr App R 23, the passage is at paragraph 211 where, without criticism, the summing‑up of the trial judge was repeated in which the trial judge said:
The important thing is this. No-one suggests that this evidence on its own conclusively proves the guilt of the defendant on any count or goes anywhere near doing that. If all you had was the DNA evidence you could not begin to find Mr Garmson guilty on any of these counts because all the DNA evidence does (at the most) is show that he is one of the men who may have committed these offences and that is perhaps to put it at its highest.
No criticism or comment was made upon the approach which was taken in the trial in that case.
BELL J: But there clearly the evidence was that the rarity was such that the accused was one of a number of persons who might have contributed the sample. That does not say anything about the admissibility or the capacity of DNA evidence, depending upon the evidence, to establish a fact.
MR HASTINGS: Your Honour we would see that as no different to this case in that it was theoretically possible that others may have come into the same profile.
BELL J: But that it the smaller proposition.
MR HASTINGS: Yes. In general, I think, your Honours, that puts our position as well as I can, unless there is anything further.
FRENCH CJ: Thank you, Mr Hastings. Yes, Mr Walker.
MR WALKER: May it please the Court. Your Honours, the importance of whether there ought to be special treatment of the kind advocated by the applicant in this case for DNA evidence in criminal cases is undeniable. There is enough material before the Court in this case and within the Court’s judicial knowledge of the widespread resort to DNA technology and statistical evidence derived from its results to make that importance great and general. In our submission, however, for the reasons I am about to summarise and will elaborate, this is not a case where the grant of special leave to appeal to this Court and the determination of an appeal thereupon will do anything, with respect, to allay the kind of concerns expressed out of court – for example, in Ms Edwards’ article to which reference has been made by the applicant in their written submissions – arising from that undeniably generally important matter.
In summary, those factors for this case largely have to do with what we respectfully submit were understandable and, for all we know, sensible decisions made by experienced defence counsel. I am going to supplement the record beyond what you have not, of course, by tendering fresh evidence, but by handing to your Honours a document which is referred to in the voir dire material which, in our respectful submission, thoroughly vindicates as commendable in his client’s interest that which defence counsel did.
In summary, that material to which I will go provided a figure, not one billion, such as the Chief Justice has used for the purposes of illustration and argument several times, but for the comparison of the reference sample and the sample from the trousers said to be derived from semen, it is designated 2A@, it is one in 20 billion, a figure which is, in relation to human population, not least in the ACT, so fantastically large as numerically to have likely an imaginative, intuitive, intellectual and rhetorical force to a jury which it might be thought, once the hurdle of admissibility had been overcome and once the question of directions had been settled, would be devastating.
BELL J: Did you say it was 20 million?
MR WALKER: Twenty billion. There is discussion of some of these numbers in the voir dire. There is another number I think, 274 billion, but I am not going to take your Honours up with that. In our submission, while of course this Court, while all criminal appellate courts, are concerned about the integrity and propriety of procedures at trial, and there is a jurisdiction which should not simply be called “residual” to overturn convictions, notwithstanding deliberate forensic choices having been made at trial by the defence, this case is a million miles away from that. This case ‑ ‑ ‑
HAYNE J: The proposition of law which you would have us establish is that there is a particular class of evidence, DNA evidence, which regardless of its particular content, should be held legally insufficient to support a conclusion that a disputed proposition of fact is established beyond reasonable doubt?
MR HASTINGS: Yes, your Honour.
HEYDON J: It is either a rule of inadmissibility or it is a rule calling for corroboration, is it not? In some circumstances it will be a rule of inadmissibility.
MR HASTINGS: Not a rule of inadmissibility, your Honour. We would say that the evidence ‑ ‑ ‑
HEYDON J: What is the point of letting in evidence that cannot do anything?
MR HASTINGS: It would be a question of adequacy, your Honour. The evidence, and we do not challenge this, has been consistently held to be admissible. The real question then, in our submission, is what force it has in any particular case and a particular case such as this where there is ‑ ‑ ‑
HEYDON J: But in your case it has no force. You say no conviction could possibly – no persuasion beyond reasonable doubt could possibly be achieved or to be permitted.
MR HASTINGS: I think I was inclined to answer earlier, your Honour, on the basis that for practical reasons that issue crystallises best at the end of the prosecution case in the no case to answer submission.
HEYDON J: In many cases that would be so.
MR HASTINGS: You may be able to do it on the voir dire in advance if the evidence was sufficiently clear but that would not always be the case. Unless there is anything further, your Honours, they are our submissions.
FRENCH CJ: Thank you Mr Hastings. Court will adjourn briefly to consider what course it should take.
AT 2.57 PM SHORT ADJOURNMENT
UPON RESUMING AT 3.21 PM:
FRENCH CJ: At trial in this matter the parties acquiesced in the statistical conclusions drawn from evidence relating to DNA profiles being expressed qualitatively rather than quantitatively. More particularly, they acquiesced in the expression of the statistical conclusions drawn from analysis of material taken from the complainant’s clothing being compared with the applicant’s DNA profile as comprising “strong” or “extremely strong” evidence in support of the contention that the applicant was the source of the material taken from the complainant’s clothing without the jury being told that the particular conclusions made by the witness in the case had yielded a figure of greater than one in 10 billion.
It was open to the jury to conclude from the evidence that was led at trial that the applicant was guilty beyond reasonable doubt. In light of the way the parties conducted the trial this is not, in our opinion, a suitable case to consider the larger question which the applicant seeks to agitate. It is the opinion of all of us that special leave should be refused.
The Court will adjourn until 10.15 tomorrow morning.
AT 3.23 PM THE MATTER WAS CONCLUDED
2