Sunday, December 22, 2019

On the evidence: the verdict in Lundy

Don’t let me fall into the trap of getting into an argument about the Lundy verdict. By verdict I mean the verdict reached by the Supreme Court justices.

I don’t think an appellate court should ever have to reach a verdict. But when it does, it can illustrate how fact-finding should rationally be undertaken.

Appellate courts usually consist of three or more judges, and differences are settled by majority vote. The defendant at a jury trial that has gone wrong may therefore end up being convicted by a majority in the appellate court. Twelve jurors or two judges – which would you prefer?

That’s the broad picture. In Lundy the Supreme Court mentioned the Court of Appeal’s point that this was a case which did not turn on the credibility of witnesses [85]. If it had been, no doubt the Court would have recognised that it was at some disadvantage because it did not hear and see the witnesses, and the disadvantage might – although not necessarily - have been such that it would not be proper for the appellate judges to try to reach a verdict.

When appellate judges reach verdicts, the right to a jury trial is limited. This was recognised at [34], where the Court refers to observations of the High Court of Australia in Weiss v R [2005] HCA 81 at [30]. Essentially, and if I might be cynical for just a moment, this means that the right to a jury trial is a right to an attempt at a lawfully conducted jury trial.

This was an entirely circumstantial case. So was Bain, and we remember how the Privy Council rejected any temptation to come to its own verdict. Even with the apparent change in approach brought about in Matenga (being more permissive of appellate court verdicts), the Privy Council said that the result in Bain would have been the same. So this whole area invites controversy.

Objections to the Lundy verdict focus on the alleged unreliability of evidence that stains on Mr Lundy’s shirt contained central nervous system tissue (“brain tissue”). Let’s put that evidence to one side, and consider what the verdict would have been without it.

The presence of Mrs Lundy’s DNA in the shirt stains was incontrovertible. The stains were not tiny: 25mm x 10mm, and 30mm x 20mm (noted at [75]). They were visible. I doubt that most people would wear [strictly on these facts, have possession of] a shirt with that staining, especially as it contained blood. And, how common is it to have blood of that quantity and from one’s partner on a shirt that has continued to be worn? Even supposing that 20% of people wear shirts with such staining – and I take this as an absurdly high occurrence in favour of the defence – this evidence strongly favours the prosecution case.

Attention then turns to the other evidence: motive, timing of food consumption, the neighbour’s observation of the open door, the fuel consumption and odometer evidence, the surprising absence of DNA evidence in Mr Lundy’s car or at the motel, the apparent staging of the burglary, the paint, and the child victim’s blood specks on Mr Lundy’s shirt. Taking all that together could firmly support the prosecution case, or weakly support the defence case, depending on the various likelihoods that fact-finders could reasonably attach to the various items of evidence.

If this other evidence in combination firmly supported the prosecution case it would be reasonable to conclude that Mr Lundy was guilty (remembering that this is ignoring the brain tissue evidence). If it weakly supported the defence case, then guilt could not be proved on the (generous to the defence) assumption that 20 per cent of people have their partner’s blood on their shirts to the extent that Mr Lundy did. If this figure is “only” 10 per cent, then guilt would be proved beyond reasonable doubt.

How do we know? This is Bayesian reasoning with conditional probabilities. Never mind if you don’t employ that somewhat mathematical approach, because common sense should tell you the same. [1]

The brain tissue evidence resulted in considerable expense and delay in the proceedings, and you could wonder whether the prosecution was over-egging its pudding. But it didn’t know whether the defence was going to have a ready answer to the DNA presence in the stains. This could be seen as a compliment to the skills of defence counsel, but we could wonder whether, in seeking to prove its case to a certainty rather than merely beyond reasonable doubt, there was (to pursue another culinary metaphor) unnecessary icing on the cake.

What if the appeal had been heard under the new appeal criteria? How would the evidence be analyzed if the court simply had to ask itself whether the error at trial had created a real risk that the outcome of the trial had been affected? The error was material to one chain of evidence: stain – CNS tissue – (mRNA) – Victim1. Taking out this mRNA chain, the remaining chain was: stain – blood – DNA – Victim 1. Did the removal of the mRNA chain significantly weaken the link between the stain and Victim 1?

[1] Oh alright, for you maths buffs: If the defendant was innocent, a huge coincidence had occurred: blood from each of the two victims was on his shirt. Such little information as there is online about transfer of blood to clothing suggests that about 3 per cent of people have blood from someone else on their clothing. This is much lower than the figures I assumed above. The ratio for the likelihood of the prosecution hypothesis to the likelihood of the defence hypothesis, the probability of finding the match given each hypothesis, would therefore be, for each of the blood sources (ie, the victims), 1 to 0.03, or 33.3 to 1. Combining these gives 33.3 x 33.3 = 1108.89 to 1. Assume, generously to the defence, that before considering this evidence we might expect that the probability of guilt compared to the probability of innocence could be, say, 0.2 to 0.8. This ratio is what is called called "the priors" combined with the other evidence in the case. On the blood evidence, and given this priors-in-combination assumption, the combined ratios are 1108.89 / 4 to 1 = 277 to 1. That is the ratio, given this evidence, of the probability of guilt to the probability of innocence. The probability of guilt would therefore be (converting the ratio to a fraction) 277 / 278 = 0.99. That is proof beyond reasonable doubt, on the basis of the assumption that was generous to the defence. Some people say, but you have just guessed at the priors, the whole thing is too subjective, where is the evidence? The answer is that the priors can vary a lot without having much effect on the result. If, instead of assuming priors combined with the other evidence of 0.2 to 0.8, we assume priors combined with the other evidence of 0.05 to 0.95 (very favourably to the defence, effectively an assumption of innocence), then the resulting probability of guilt after considering the blood evidence is 0.98. That is: (0.05 / 0.95) x 1108.89 = 58.36 to 1, so probability of guilt = 58.36 / 59.36 = 0.98.

Friday, December 20, 2019

Tempora mutantur nos et mutamur in illis - or do we?

Would the result in Lundy v R [2019] NZSC 152 have been the same if it had been determined under our current appeal provisions, s 232 of the Criminal Procedure Act 2011?

Applying the former law, the Supreme Court held that the wrongful admission of evidence at trial had not created a substantial miscarriage of justice because if the error had not occurred guilt would still have been proved beyond reasonable doubt, and the trial had not been unfair because the error was not fundamental. It was not necessary to consider the new law, or the extent to which the old law might be affected by the legislative change. But this is really the live issue for lawyers.

If the new law had applied to the Lundy appeal, s 232(4)(a) would be of central concern: did the error at trial create “a real risk” that the outcome of the trial was affected?

It is helpful to consider some dicta from the High Court of Australia in Baini v The Queen [2012] HCA 59. I have previously discussed this case here (December 13, 2012).  As a matter of interpretation, the Court had to ask what was the relevance of a former appeal provision (similar to our former one) to ascertaining the meaning of a new appeal provision. The new legislation considered in Baini is set out at [12], and although its form and language differs from ours, the difference is not presently material. The approach to its construction is described at [14]-[15]. The majority judges then made three observations:

At [29]:

“First, in many cases ... an appellate court will not be in a position to decide whether the appellant must have been convicted if the error had not been made. The nature of the error, irregularity or cause of complaint contemplated by [the legislation] will often prevent that conclusion from being reached by an appellate court on the record of the trial given the "natural limitations" [Footnote: See Dearman v Dearman (1908) 7 CLR 549 at 561; [1908] HCA 84; Fox v Percy (2003) 214 CLR 118 at 125-126 [23]; [2003] HCA 22.] that attend the appellate task.”

At [31]:

“If it is submitted that a guilty verdict was inevitable, an appellant need not prove his or her innocence to meet the point. An appellant will meet the point by showing no more than that, had there been no error, the jury may have entertained a doubt as to his or her guilt. As a practical matter, it will then be for the respondent to the appeal to articulate the reasoning by which it is sought to show that the appellant's conviction was inevitable.”

And at [32]:

“... the inquiry to be made is whether a guilty verdict was inevitable, not whether a guilty verdict was open. (Whether the verdict was open is the question presented by [the equivalent of New Zealand’s s 232(2)(a)].) If it is said that a guilty verdict was
inevitable (which is to say a verdict of acquittal was not open), the Court of
Appeal must decide that question on the written record of the trial with "the
'natural limitations' that exist in the case of any appellate court proceeding wholly or substantially on the record” [Footnote: Fox v Percy (2003) 214 CLR 118 at 125-126 [23]]. That the jury returned a guilty verdict may, in appropriate cases [Footnote: See generally Weiss (2005) 224 CLR 300 at 317 [43]; Baiada Poultry Pty Ltd v The Queen (2012) 86 ALJR 459 at 466 [28]; 286 ALR 421 at 430; [2012] HCA 14.], bear upon the question. But, at least in cases like the present where evidence has wrongly been admitted at trial and cases where evidence has wrongly been excluded, the Court of Appeal could not fail to be satisfied that there has been a substantial miscarriage of justice unless it determines that it was not open to the jury to entertain a doubt as to guilt [Footnote: cf R v Grills (1910) 11 CLR 400 at 431 per Isaacs J; [1910] HCA 68.] Otherwise, there has been a substantial miscarriage of justice because the result of the trial may have been different (because the state of the evidence before the jury would have been different) had the error not been made.”

And at [33]:

“ appellate court can only be satisfied, on the record of the trial, that an error of the kind which occurred in this case did not amount to a "... miscarriage of justice" if the appellate court concludes from its review of the record that conviction was inevitable. It is the inevitability of conviction which will sometimes warrant the conclusion that there has not been a ... miscarriage of justice with the consequential obligation to allow the appeal and either order a new trial or enter a verdict of acquittal.”

There would be no “real risk” that the outcome of the trial had been affected by the error if the appeal court could be satisfied that the verdict was inevitable. That was the position in Lundy. The result would have been the same under our new law. On this view, our new law would continue the practice of sometimes allowing appeal courts to effectively act in place of a jury. It will be important to identify whether a case falls into the category mentioned at [29] of Baini, as opposed to the category mentioned at [31], and to decide whether this classification is determined by the respondent’s tactic in argument.

But the other view, and the one which I prefer, is (developing what is said at [31] above) that once the appellant shows a real risk that the verdict was adversely affected by the error, it is for the respondent to dispute that; the issue is not inevitability, just real risk. The appeal judges do not act as jurors; they just decide the real risk issue.

Update: A year later, on 18 December 2020, the New Zealand Supreme Court decided how appeals against conviction based on alleged error relating to a trial (ie not those involving allegedly unreasonable verdicts, unfairness or nullity) should be approached. Effectively, the Court read-in a proviso: if an error that is not immaterial or inconsequential is identified, and if it gave rise to a real risk that the result of the trial was affected, then the appeal court must decide whether it is satisfied of the appellant's guilt. The appeal will be allowed, provided (my word) that it will be dismissed if the appeal court is satisfied of the appellant's guilt: [Name Suppressed] v R [2020] NZSC 153. This is not the approach I have preferred. It amounts to a trial by appellate judges, in the appellant's absence, without witnesses appearing, and in reliance on the record of the trial in the lower court.

Friday, December 13, 2019

All in the way you look at it

De Silva v The Queen [2019] HCA 48 (13 December 2019) is one of those cases in which, for an uninvolved reader, the dissenting judgment is just as persuasive as the majority’s. Not to worry, no doubt you had to be there.

In determining whether the judge’s directions to the jury had been wrong, the majority pointed to the absence at trial of any request for clarification of them from defence counsel (at [35]).

I can’t help wondering whether it is overly ambitious to place weight on counsel’s failure to point to a perceived error. Counsel, being well versed in the law, will tend to hear the judge say what the judge means to say, even if the judge hasn’t said it. When counsel knows what the judge means, it is very difficult to misunderstand what the judge says in the way that a jury may misunderstand it.

Only after the event, when there has been time for considered reflection on how misunderstandings may have arisen, might errors become apparent.

Still, the case has some useful general points. Nettle J (dissenting in the result) mentions the standard direction on beyond reasonable doubt (at [47]). The majority emphasise that directions to juries should not include unnecessary matters (here, whether a statement was made, when it was obvious from the recorded interview that the statement was made) [33], and a Liberato direction (in NZ we call this a tripartite direction [1]) can, depending on the issues and the conduct of the trial, be given even if the defendant’s version of events only comes into evidence via a statement given to the police [13].


[1] R v McI [1998] 1 NZLR 696 (CA) at 708, cited in R v MacDonald [2009] NZCA 428, (2009) 24 CRNZ 785 at [2], and Hazelwood v R [2013] NZCA 406 at [47].

Friday, December 06, 2019

New hardcopy edition of Misuse of Drugs

As any intelligent person knows, it is better to read about drug offences than to commit them.

With that in mind, it can only be helpful for me to point out that the new hardcopy edition of my textbook Misuse of Drugs can now be ordered here.

Wednesday, December 04, 2019


If you ever thought statutory interpretation could sometimes be a bit puzzling, take comfort from the High Court of Australia’s struggle in New South Wales v Robinson [2019] HCA 46 (4 December 2019). The Court split 4-3, and did well to limit itself to two judgments.

Of more general interest, something to add to your collection of dicta explaining what “reasonable grounds to suspect” means, see [55] of the minority judgment.

Tuesday, December 03, 2019

As Clive James said ...

Now that rain has stopped play at Seddon Park, let me share some of my favourite Clive James quotes:

From The Meaning of Recognition – New Essays 2001-2005:

“ ... she was talking too fast to hear herself think.”

“... there is a lingering aesthetic sense that appreciates a tall, slim female form draped in a black linen pants suit underpinned with strappy high-heeled sandals for the shapely feet, the toenails painted with the blood of slain lovers.”

“The concept was too subtle for her to grasp. It turned out that she was a lawyer. She can attack me as often as she likes, but I hope to God that she never defends me.”

“Those of us who have lost our hair find it hard to suppress the suspicion that it has been stolen in the night by those who have kept theirs.”

“... I found Moby Dick a hard read ...”.

“... academic politics in all their bitchy intricacy ...”.

“... you have to admire a country so democratic that a mentally handicapped man can become President.”

“As all you women in the audience know, the amount of hair on top of a mature man’s head is governed by the amount of testosterone he secretes, but the proportion is not direct. The proportion is inverse. Testosterone attacks the hair follicles. It fries and shrivels them like noodles in a wok of acid.”

From Cultural Amnesia:

“One famous compulsive fabulist, the Jeffrey Archer of his time, never recovered what was left of his credibility after Moln├ír said: ‘He’s such a liar that not even the opposite is true’.”

“As I write, the elite is in its last phase, where it begins to forget the car keys through the effort of remembering the door keys.”

“The hardest trick in an Italian poet’s book is to avoid rhyme ...”.

“He was the sort of writer who said ‘microphone’ when he meant ‘loudspeaker.’”

“... I became a scolding bore on the subject of the Cinquecento. On the Quattrocento, less hindered by academic assistance, I was capable of the odd independent judgement ...”.

“... a shock, as if Jane Austin had forgotten to mention money.”

“Written in an age when it was still assumed that children would not suffer brain damage from hearing a phrase they couldn’t immediately understand ...”.

“... the flouncing kick-line of the post-modern intellectual cabaret.”

“Expressing oneself clearly is the most complicated thing there is.”

“Like literary theory at a later time, however, analytical philosophy was a hard game to get out of after you had started drawing the salary.”

Monday, December 02, 2019

Appellate review to identify miscarriages of justice in conviction appeals

If you were ever worried about whether conviction appeals under s 232(2)(c) and (4) of the Criminal Procedure Act 2011 are determined by way of review or by rehearing, relax. Just by looking at Misa v R [2019] NZSC 134 you can see that the appeal court does not try to reach its own verdict (as it would if an appeal were to be heard by way of rehearing), so the approach is by way of review.

Just a wee cautionary note: don’t be confused by s 78 of the Senior Courts Act 2016, which provides that appeals to the Supreme Court are heard by way of rehearing. It just means a rehearing of the appeal in the court below, which was by way of review: see Sena v Police [2019] NZSC 55 (discussed here) at [25, fn 41] (referring to an identical earlier provision).

True, the Court in Misa does not declare itself to be conducting a review of what happened at the trial, but its focus was on substance rather than terminology. In substance ([48]): “the question is whether the error, irregularity, or occurrence in or in relation to or affecting this trial has created a real risk the outcome was affected. That, in turn, requires consideration of whether there is a reasonable possibility another verdict would have been reached.”

The second sentence elaborates the first which is a repetition of the statutory language of s 232(4)(a). That is, “real risk” equals “reasonable possibility” and “was” equals “would”.

On the facts, the proposed new evidence would not have made any material difference, so there was no error, irregularity or occurrence in or in relation to or affecting the trial that created a real risk that the outcome was affected (at [77]).

This conclusion was reached, not by the Court considering the whole of the evidence and reaching its own verdict, but by identifying the defence strategy at trial and assessing the effect of the proposed new evidence on that (at [74]). The defence strategy was to allege collusion as an explanation of the similarity of the evidence of two prosecution witnesses, thereby undermining them both. It was not to create differences between the witnesses. In the circumstances the chosen defence strategy was sound, and there had been no error of trial counsel in using it. Further, the proposed new evidence was not inconsistent with evidence that had been given at trial and it raised no new issues (see [76] and its references to the jury’s perspective).

But, you may say, won’t the Crown usually oppose the appeal against conviction by arguing that the conviction was inevitable on the evidence, so that to address that argument the appellate court will have to come to its own verdict? In other words, rehear the case on the record of the trial? That this is not precluded was recognised in Wiley v R [2016] NZCA 28 at [44], [49].

The Court in Misa did not have to address this aspect of Wiley, which was upheld in other respects (at [36]-[37]). Instead of arguing that the verdict was inevitable, the Crown here submitted, successfully, that the new evidence was inconsequential.

Thursday, November 14, 2019

On being helpful

A defendant’s right to a fair trial is an absolute right, but is the right to procedural fairness absolute? Can a trial be fair if its procedure is less than fair?

At sentencing, the defendant may claim mitigation for having given assistance to the authorities. There may be good reasons for not making public the fact that such assistance was given, and procedures may be in place to provide information to the judge on a confidential basis. Such information can take the form of a memorandum prepared and agreed to by counsel for each side. But what if the prosecutor wishes to include information for the judge’s eyes only, favourable to the defendant, but where disclosure of it to the defendant might compromise ongoing investigations?

This was considered in HT v The Queen [2019] HCA 40 (13 November 2019). Three judgments were delivered, all agreeing in the result. The majority opinions were given jointly by Kiefel CJ, Bell and Keane JJ, with support from Gordon J. Generally, where this problem arises, counsel should approach the sentencing judge for an appropriate order restricting disclosure, for example to counsel for the defendant but not to the defendant, or for disclosure to independent counsel who will then obtain instructions from the defendant to the extent that it is possible to do so while still keeping the confidential information from the defendant. At the sentencing hearing suppression and non-publication orders, or an order for an in camera hearing, could also be made.

Nettle and Edelman JJ accepted that suppression and in camera orders could be used, but they dissented by holding that the defendant must see everything that the judge sees [59]. It should only rarely not be possible for counsel to agree on a memorandum describing assistance given by the defendant, and the drafting of a suitable memorandum must be done in the light of the prosecutor’s duty of fairness. They pointed to the need for legislation to cover situations where agreement could not be reached.

In New Zealand it has been held that there is no power to exclude the public from a sentencing hearing, or to hear submissions in chambers. There are rules concerning assistance to the authorities: Criminal Procedure Rules 2012, r 5A.8. Sentencing involves deciding on a starting point in view of the seriousness of the offending and the role of the offender, and then adjusting it for mitigating or aggravating circumstances relating to the offender, and then giving credit for any assistance rendered by the offender to the authorities, and finally giving such credit as is appropriate in the circumstances for any guilty plea. In the result, the mitigatory effect of assistance to the authorities can be substantial. Failure to follow the sealed envelope procedure in the rules led to problems in AB(CA300/2017) v R[2018] NZCA 51 (not publicly available online), where on appeal credit by way of reduction of sentence by one third was considered appropriate, in the absence of a guilty plea, but where important information had been given in relation to serious offending. Orders were made suppressing the details of the assistance, the identifying particulars of the appellant, and limiting access to the court file. The Court noted that there have been cases where, when credit for assistance is combined with that for a guilty plea, a reduction from the appropriate sentence of up to 60 per cent has been approved.

Thursday, November 07, 2019

An unreasonable jury verdict

The criminal courtroom can make innocence look like guilt.

A combination of circumstances persuaded a jury and a first appeal court that a murder had been committed by the defendant, but on second appeal the verdict was held to have been unreasonable, the conviction was quashed and an acquittal entered: Fennell v The Queen [2019] HCA 37 (6 November 2019).

It is worth reading the judgment for an illustration of how to take apart a prosecution case which relies on circumstantial evidence.

From the point of view of the law on how an appeal court may decide whether a jury verdict is reasonable, see [81]-[82]:

“[81] Where a court of criminal appeal is called upon to decide whether it considers that, upon the whole of the evidence, it was open to the jury to be satisfied beyond reasonable doubt that the accused was guilty of the offence charged, the court must not disregard or discount either that the jury is the body entrusted with primary responsibility of determining whether the prosecution has established the accused's guilt or that the jury has had the benefit of having seen and heard the witnesses. [M v The Queen (1994) 181 CLR 487 at 493.] At the same time, however, the court may take into account the realities of human experience, including the fallibility and plasticity of memory especially as time passes, the possibility of contamination of recollection, and the influence of internal biases on memory. [Bingham, "The Judge as Juror: the Judicial Determination of Factual Issues" (1985) 38 Current Legal Problems 1 at 16-18.] The court can also take into account the well-known scientific research that has revealed the difficulties and inaccuracies involved in assessing credibility and reliability. [Fox v Percy (2003) 214 CLR 118 at 129 [31]. See also Devlin, The Judge (1979) at 63; Bingham, "The Judge as Juror: the Judicial Determination of Factual Issues" (1985) 38 Current Legal Problems 1 at 7-8, 10-11; R (G) v Governors of X School [2012] 1 AC 167 at 196-197 [80].] And especially is that so in a case like this where the jury has been subjected to the seductive effects of a species of identification evidence that has in the past led to miscarriages of justice. [See R v Clout (1995) 41 NSWLR 312. See and compare Domican v The Queen (1992) 173 CLR 555 at 561.] ... .

“[82] In an appeal where a ground of appeal is that the verdict was unreasonable or cannot be supported having regard to the evidence, and particularly where the Crown case is based upon a number of matters of circumstantial evidence, it is necessary for the appellate court to assess the whole of the case and to weigh that case as a whole. A circumstantial case cannot be considered in a piecemeal fashion. [R v Hillier (2007) 228 CLR 618 at 638 [48]. And see my note on this case here.] A corollary of this principle, particularly in cases in which the conclusion is not clear-cut, is that a substantial error, or errors, by an intermediate court of appeal in the process of assessing the case as a whole can infect the entire process.”

Here, the first appeal court had made an important error. The credibility of a significant item of evidence (an assertion that a hammer had been lent to the defendant) was greatly over-estimated ([76]-[80], [89]). This error may, I suspect, have occurred because, if true, the evidence would have been highly probative. The treatment of other circumstances as also being more sinister than they were could have led the court to over-estimate the likelihood that the hammer had been correctly identified.

Generally speaking, two steps are involved: first, assessing the credibility - or reliability [1] - of the witness's testimony ("This is the hammer I lent the defendant"), and second, if it is sufficiently credible to satisfy the fact-finder that it is true, the assessment of its probative value on an issue in the case. There is not necessarily a standard of proof for deciding whether testimony is credible, although it is often said that if the evidence must be accepted before a verdict of guilty can be returned then the credibility of that item of evidence must be established beyond reasonable doubt. For less central testimony it is likely that credibility need only be established to a level that "satisfies" the fact-finder that it is true, and here satisfies means on the balance of probabilities. However, this is a rather vexed topic and jurisdictions may differ in how they express the requirement for acceptance of the truth of evidence.

Credibility and probative value are matters specific to the particular testimony that is being considered, but when it comes to assessing the verdict on the accepted evidence, it is all the accepted evidence that must be taken into account. This is why the court can say it is wrong to latch on to one item of evidence and claim that it creates a reasonable doubt about the defendant's guilt. There will be occasions where that can be done, for example if an alibi is accepted notwithstanding that the other evidence strongly supports an inference of guilt. But other than alibi evidence, and also evidence supporting an affirmative defence (where the defendant's actus reus and mens rea are not disputed), in most cases, including Fennell, all the accepted evidence must be considered in deciding whether guilt has been proved beyond reasonable doubt.

In the course of a criminal trial, as evidence of guilt appears to accumulate, further evidence may be given greater credibility than it deserves.

But that’s not all. References by the High Court of Australia in Fennell to Bingham remind us of how Lord Bingham, delivering the judgment of the Privy Council, analysed our notorious circumstantial evidence case, Bain v R [2007] UKPC 33 (10 May 2007), noted here. But the Bain appeal was different insofar as it involved fresh evidence which called into question whether the combined effect of the circumstantial evidence would, in its light, be sufficient to support a verdict of guilty. As there was fresh evidence, it was appropriate to order a retrial. It was not for the appellate court to itself come to a verdict on the new evidence. In contrast, Fennell was not a fresh evidence case. The prosecution relied on categories of evidence - alleged theft, alleged motive, alleged possession of the murder weapon, and alleged opportunity - and each was shown to be unreliable. This was a case of absence of reliable evidence, not a case of assessing on an overall view the probative value of reliable evidence. On finding the conviction unsound the Court could order an acquittal.

[1] In New Zealand s 28 of the Evidence Act 2006 is relevant where the reliability of a defendant's statement is being considered, and this has been interpreted as meaning that the truthfulness of the evidence can be relevant to assessing its reliability, a position not criticised by the Law Commission in its Second Review of the Evidence Act 2006NZLC R142 at [6.19]-[6.23]. In R v Wichman [2015] NZSC 198 at [76] the majority in its joint judgment observed in relation to the possibilities that someone who is innocent may be convicted on evidence which turns out not to have been reliable, that these risks "... are inherent in the criminal trial process and cannot be completely avoided. They can, however, be mitigated in various ways; in particular by recognising the risk of unreliability; the provision of a reliability screening process to be carried out by the judge; judicial warnings; jury assessment and the rules as to the burden and standard of proof."

Wednesday, October 23, 2019

Sentencing for methamphetamine dealing

In a splendidly clear judgment our Court of Appeal has revised the law on sentencing for methamphetamine dealing (importing, manufacturing, and supplying): Zhang v R [2019] NZCA 507.

The Court of five Judges helpfully sets out a summary of the judgment at [10], and adds at [11]:

“...Those who willingly participate in commercial-level dealing in methamphetamine will gain little succour from this judgment. Its benefits lie more for those who take a lesser role in methamphetamine offending, and particularly those who do so as a result of vulnerability.”

There is no need for me to summarise the case (six jointly heard appeals), as the Court has done the hard summarising work in [10] and [11].

I make two observations.

First: If you thought that in referring to only two stages of analysis the Court was overlooking the taking into account of an early guilty plea, don’t worry. Looking at the discussions of the cases appealed you can see, for example at [201] and [258] the traditional approach.

Second: You may remain confused over the way an appropriate sentence for a conspiracy is identified. At [239] the Court says that taking half the penalty identified for the full offence is the correct approach, citing in footnote 133 Parata v R [2017] NZCA 48 at [5]. Yes, you say, but in Parata no authority for that was given. I dare not suggest that there might be some confusion here between s 310(1) of the Crimes Act 1961 and s 311(1) of that Act which provides (materially) for half the penalty for the full offence if only the attempt is committed. But s 310 does not apply here either, because the penalties for conspiracy to deal in drugs are set out in s 6(2A) of the Misuse of Drugs Act 1975. They are 14, 10 and 7 years’ imprisonment, compared with, for the corresponding full offences, life, 14 and 8 years. The approach to sentencing in the drugs context for conspiracies has been to reduce the levels set out in the bands by about 30 per cent: R v Haarhaus HC Auckland CRI-2007-004-018646, 24 July 2009, applied in R v Feng [2016] NZHC 1333 (neither of these cases was referred to in Zhang). Nor did the Court refer to its earlier decision in R v Te Rure [2007] NZCA 305, [2008] 3 NZLR 627, where detailed consideration was given to conspiracy sentencing in the context of dealing guidelines. So, what should we do? Follow High Court authority which has not been overruled, and take (as the first step in setting a starting point) about 30 per cent off the penalty identified in the new bands if the full offence had been committed? Or follow Zhang which, for no apparent reason, halves the penalty for the full offence as identified in the bands?

You decide.

New edition of Criminal Procedure in New Zealand

Don’t let me stop you ordering your copy of the newly published Criminal Procedure in New Zealand (3rd ed) by Emeritus Professor Jeremy Finn and me.

Tuesday, October 08, 2019

Probative value and prejudicial effect: when weighing is not balancing

Weighing probative value against unfairly prejudicial effect
Section 8 of the Evidence Act 2006 sets out a generally applicable exclusionary rule:
8 General exclusion
(1) In any proceeding, the Judge must exclude evidence if its probative value is outweighed by the risk that the evidence will—
(a) have an unfairly prejudicial effect on the proceeding; or
(b) needlessly prolong the proceeding.
(2) In determining whether the probative value of evidence is outweighed by the risk that the evidence will have an unfairly prejudicial effect on a criminal proceeding, the Judge must take into account the right of the defendant to offer an effective defence.
The weighing exercise in subsection (1) was introduced into our evidence legislation by the Evidence Amendment Act (No 2) 1980. This stated in s 18:
18 Court may reject unduly prejudicial evidence
Notwithstanding sections 3 to 8 of this Act [the Evidence Act 1908], where the proceeding is with a jury, the Court may, in its discretion, reject any statement that would be admissible in the proceeding under any of those sections, if the prejudicial effect of the admission of the statement would outweigh its probative value, or if, for any other reason, the Court is satisfied that it is not necessary or expedient in the interests of justice to admit the statement.
Obvious differences are the elimination of the description of the weighing exercise as a discretion, and the introduction of the word “unfairly” to qualify “prejudicial effect”. Another change is the omission of the reference to the interests of justice.
The Law Commission’s initial comments
In relation to the current provision, the Law Commission said of probative value:[1]
Probative value will depend on such matters as how strongly the evidence
points to the inference it is said to support, and how important the evidence
is to the ultimate issues in the trial.
As to unfair prejudice, the Commission said:[2]
There must be an undue tendency to influence a decision on an improper or illogical basis, commonly an emotional one; for instance, graphic photographs of a murder victim when the nature of the injuries is not in issue. Evidence will also be unfairly prejudicial if it is likely to mislead the jury; for example, if it appears far more persuasive than it really is, as is occasionally the case with some types of expert and statistical evidence. The judge will need to consider whether any misleading tendency can be countered by other evidence that is likely to be available, or by a suitable direction to the jury. Whether evidence has an unfairly prejudicial effect must be considered in terms of the proceeding as a whole, and not just from the point of view of a particular party or a defendant.

A similar weighing exercise applies to propensity evidence pursuant to s 43(1) of the Evidence Act 2006. This was brought to attention recently in Brooks v R,[3] where, in refusing leave to appeal, the Supreme Court quoted from its earlier decision, the leading authority:
“[8] When a judge is considering the extent of any unfair prejudicial effect on the defendant [under s 43], the judge should examine whether the fact that the propensity evidence is prior acquittal evidence gives rise to any, or any additional, unfair prejudice. To the extent that it does, the judge should consider how that additional dimension affects the overall balance between probative value and unfair prejudice ...”. [emphasis added]
It is this “overall balance” that is, I suggest, a misuse of the metaphor of weighing. The section uses the word “outweighs”, and the only rational interpretation in context is that it means that the risk of an unfairly prejudicial effect on the defendant is negligible; it is contextually insignificant insofar as there is no real risk of trial unfairness. Unfortunately, outweigh also means to be greater or more significant than, and this tempts the reader to invoke a balancing metaphor. Although this weighing exercise is cemented in legislation, and therefore must be applied when it is relevant, it is appropriate to note that, when understood as necessitating balancing, it has been criticised as requiring a comparison of things that are not comparable. While it is easy to accept that probative value can have metaphorical weight, unfair prejudice is something that either exists or it doesn’t, and it is not apt to think of it as having weight.[4]
Common law origins
The origins of this weighing exercise are in the common law, as applied to New Zealand indirectly via the Privy Council’s decision in Noor Mohamed v R (British Guiana):[5]
“... the judge ought to consider whether the evidence which it is proposed to adduce is sufficiently substantial, having regard to the purpose to which it is professedly directed, to make it desirable in the interests of justice that it should be admitted. If, so far as that purpose is concerned, it can in the circumstances of the case have only trifling weight, the judge will be right to exclude it. To say this is not to confuse weight with admissibility. The distinction is plain, but cases must occur in which it would be unjust to admit evidence of a character gravely prejudicial to the accused even though there may be some tenuous ground for holding it technically admissible. The decision must then be left to the discretion and the sense of fairness of the judge.”
This is not a balancing exercise. It reflects what had been said in R v Christie,[6] about there being a “practice of a very salutary nature” that had “long existed” (although there is apparently no confirmation of that in the law reports)[7]of the judge inviting the prosecutor not to persist in adducing evidence that, although legally admissible, would operate unfairly against the defendant. Lord Moulton said:
“Under the influence of this practice, which is based on an anxiety to secure for everyone a fair trial, there has grown up a custom of not admitting certain kinds of evidence which is so consistently followed that it almost amounts to a rule of procedure.”
Christie is often referred to as the source of this exclusionary common law evaluative judgement, but it was not expressed there as a weighing exercise. Nevertheless, evidence legislation internationally often adopts the weighing formula, although the United Kingdom legislation now avoids a weighing model in favour of a requirement that evidence must not be admitted if admission would have “such an adverse effect on the fairness of the proceedings that the court ought not to admit it.”[8]
The Law Commission’s 2008 comments
Our Law Commission has, in a report in 2008, had more to say about this:[9]
“7.55 The so called “weighing” of probative value and risk of prejudicial effect required by these provisions needs some elucidation. It is something of a shorthand expression, picked up and carried forward from the common law into the statute. As a matter of logic, probative force and prejudice are incommensurables. The comparison is not one of like with like. What really occurs is that the judge evaluates the probative strength in the trial of the evidence, and also the degree of risk of an unfair trial through prejudice if it is admitted. The effect of each on the fair trial process then is evaluated, and a value judgement is made based on the interests of justice....

“7.56 ...What matters in the “weighing” exercise is whether the evidence carries a risk of prejudice relative to probative value which goes beyond the bounds of the legitimate by creating an unacceptable degree of risk of an unfair trial. ... will the propensity evidence be given greater force than it could properly warrant, bolstering an otherwise weak Crown case? It is the risk of “illegitimate” prejudice which is the concern; and at which the weighing exercise is directed.”

This amounts to acknowledging that the legislation is not to be applied literally. It is “so called” weighing, needing “some elucidation”, it is “something of a shorthand expression”, “carried forward from the common law”. The Commission seems to be wishing that courts could “elucidate” the legislation in the same way they can alter the common law by explaining it. This explanation of the weighing exercise (“what really occurs”) amounts to a new model for the decision process. Although perhaps acceptable – but that may be being generous - for the time in which it was made, it needs to accommodate the recognition in the Supreme Court of decisions which the Commission does not cite, upholding the absolute nature of the right of the defendant to a fair trial.[10]

The Law Commission’s 2013 Review
When it reviewed the working of the Evidence Act in 2013, the Law Commission appears not to have received submissions on whether the balancing model should be retained or amended.[11] The issues relating to propensity evidence that were considered were in effect a repetition of the discussions in Mahomed v R.[12] The Commission did not criticise the Court’s view that there is little difference between the “balancing” exercises in s 8 and s 43(1),[13]while noting that there is “at least [a] theoretical” difference if the balance is determined to be equally poised. Importantly, the Commission concluded that it was not prepared to countenance any approach that would give rise to a risk of unfair trials, and this governing principle led it to reject submissions that were directed at linking (by a so-called “tit for tat” rule) the permitting of a defendant to cross-examine a complainant in a sex case about previous sexual history with permission to lead evidence of a defendant’s past sex offence convictions.[14]
In dealing with the hypothetical example of the balance being “equally poised” while at the same time upholding the need to avoid giving rise to a risk of unfair trials, the Commission overlooked the level of risk to trial fairness posed by an equal balance. A likelihood of unfairness of 50% would be regarded as too great a risk of unfairness. The decision should be exclusion when the likelihood of unfairness is much lower. The hypothetical example really asked what happens when the balance is on the boundary between admission being fair and not fair.[15]
In assessing this risk of unfairness the judge will have regard to the likely efficacy of any instruction that the jury may be given about the proper use of the evidence and any warning about avoiding its improper use.
Prior acquittal evidence
Acquittal does not mean innocence. It could be consistent with proof of guilt but only to the insufficient standard of the balance of probabilities. When finding facts in a criminal trial, the balance of probabilities will be sufficient in deciding what evidence to accept and what to reject. Of course, the combined effect of the evidence that is accepted is the basis for the verdict, and a verdict of guilty requires the combined effect to be proof beyond reasonable doubt.
This means that evidence given at a defendant’s earlier trial where the verdict was not guilty can be adduced if it is relevant as propensity evidence in a subsequent trial for a different offence. This does not infringe the rule against double jeopardy because at the subsequent trial the defendant is not in jeopardy of conviction for the conduct that was the subject of the earlier trial. Brooks and Fenemor (above) were cases of this sort. The focus is on what the defendant claims to be prejudice that is unfair: there must be “something about the circumstances of or leading to the acquittal which gives rise to prejudice that is unfair”.[16]
The split in the Court of Appeal[17] in Brooks over whether there was unfair prejudice, is an indication of how difficult this evaluative judgement can be. The majority, Miller and Simon France JJ, said this about unfair prejudice in this context, and held that none had been demonstrated:
“[41] Following Fenemor, a defendant may be able to point to something about a previous acquittal which makes the admission of evidence tending to prove the same allegation unfairly prejudicial for purposes of s 43. That may arise where admission of the evidence forces the defendant to adduce evidence showing how the acquittal came about.
“[42] By way of illustration, the defendant may have offered a successful alibi, as noted above, and will need to re-establish it. Defence witnesses may now be unavailable. The record of the first trial may establish that the evidence was seriously deficient in some way, [citing RPG v R 2015] NZCA 275] or a key witness lacked credibility. [citing Blackburn v R [2011] NZCA 365 and R v Edwards [1991] 2 All ER 266 (Crim App)]  (If that witness is to give the propensity evidence, the evidence may also be of low probative value for s 43 purposes.) The defendant may also show that the need to call evidence bearing on the acquittal contributes to a risk that the propensity evidence will overwhelm the trial, unduly influencing the fact-finder.
“[43] We accept that in such a case unfair prejudice may result where a police or trial file is unavailable and the relevant evidence cannot be established in another way. That conclusion is consistent with the authorities. [...]  However, following Fenemor , the court will not assume that the mere absence of a file is prejudicial. The defendant must both point to something about the acquittal that is relevant and show that the absence of the file is prejudicial.”
However, Peters J was unable to agree that there was no unfair prejudice in this case. Effective cross-examination was not possible because of passage of time and the absence of a proper record:
“[58] First, there is the matter of delay and its adverse effects on the accuracy of memory [CT (SC 88/2013) v R [2014] NZSC 155, [2015] 1 NZLR 465 at [14] and [26]; and Rompa v R [2010] NZCA 277 at [11].] The fact that this Court noted Mr Brooks' inability to recall details of his earlier trial is significant to me.

“[59] Secondly, almost all relevant contemporaneous information was unavailable, that is, the information from the police investigation and trial. Of course, this could occur in respect of any documents relevant to propensity evidence, but in this case it comprised the notes of evidence, police job sheets, notebook entries and (subject to what is said below) witness statements and/or briefs of evidence. [The Detective], the officer in charge of J's trial, gave evidence that these documents would have existed, but could not now be found or that (former) [another detective], the officer in charge of the S trial, had advised that the documents had been destroyed. Also, the video of Mr Brooks' own police interview was unavailable and a transcript of it which appeared to have been prepared was also unavailable. In fact, the only available contemporaneous information of consequence comprised the witness statements and briefs of evidence of S and a related witness, D. The Crown provided these documents to S and D prior to trial to refresh their memory. Even then, each was unable to recall matters of detail when giving evidence.”
At least there was no suggestion here that if propensity evidence had a high probative value it could only be outweighed by prejudice that had an even greater value. Unfairly prejudicial evidence is evidence which gives rise to an unacceptable risk of trial unfairness, regardless of how much probative value it may have.

[1]Evidence Code and Commentary (NZLC R55 Vol 2 1999) at C58.
[2]Evidence Code and Commentary (NZLC R55 Vol 2 1999) at C59.
[3] Brooks v R [2019] NZSC 107 (8 October 2019), quoting Fenemor v R [2011] NZSC 127 at [8].
[4]Compare McHugh J in Pfennig v R [1995] HCA 7, (1995) 182 CLR 461 at 528, [39], almost making the same point: “...The use of the term ‘outweigh’ suggests an almost arithmetical computation. But prejudicial effect and probative value are incommensurables. They have no standard of comparison. The probative value of the evidence goes to proof of an issue, the prejudicial effect to the fairness of the trial. In criminal trials, the prejudicial effect of evidence is not concerned with the cogency of its proof but with the risk that the jury will use the evidence or be affected by it in a way that the law does not permit. In no sense does the probative value of evidence disclosing propensity, when admitted, outweigh its prejudicial effect. On the contrary, in many cases the probative value either creates or reinforces the prejudicial effect of the evidence. In my view, evidence that discloses the criminal or discreditable propensity of the accused is admitted not because its probative value outweighs its prejudicial effect but because the interests of justice (181 cf. Lord Morris in Boardman (1975) AC at 439) require its admission despite the risk, or in some cases the inevitability, that the fair trial of the charge will be prejudiced.” The reason given in this last sentence is now not correct if the defendant’s right to a fair trial is absolute.
[5]Noor Mohamed v R [1949] AC 182, 192.
[6]R v Christie [1914] AC 545.
[7]MS Weinberg, “The Judicial Discretion to Exclude Admissible Evidence” (1975) 21 McGill Law Journal 1 at 23.
[8]Criminal Justice Act 2003 [UK], s 101(3), Police and Criminal Evidence Act 1985 [UK], s 78.
[9]Disclosure to Court of Defendants’ Previous Convictions, Similar Offending, and Bad Character NZLC R103 at [7.55]-[7.59].
[10]Condon v R [2006] NZSC 62, [2007] 1 NZLR 300 at [77]. The Report was published in May 2008 yet does not cite Condon. Subsequently in Mohamed v R [2011] NZSC 52, [2011] 3 NZLR 145, (2011) 25 CRNZ 223 at [66] per McGrath and Tipping JJ it was said that under s 43(1) ‘the only relevant prejudicial effect which logically needs to be considered is the likely impact on the fairness of the trial from the point of view of the defendant.”
[11]The 2013 Review of the Evidence Act 2006NZLC R127.
[12]Mahomed v R [2011] NZSC 52, [2011] 3 NZLR 145, applied in Turner v R [2014] NZCA 480.
[13]The 2013 Review of the Evidence Act 2006NZLC R127 at [6.120], [6.126], [6.133].
[14]The 2013 Review of the Evidence Act 2006NZLC R127 at [6.136]-[6.137].
[15]Aside from that point, the Commission agreed that if this theoretical situation were to occur, s 8 would require admission (as the probative value was not outweighed), while for propensity evidence s 43(1) would require exclusion (as the probative value did not outweigh the risk of unfairness).
[16] Brooks v R [2019] NZSC 107 at [6].
[17] Brooks v R [2019] NZCA 280 (the post-trial CA judgment, not currently freely available online).