Beyond the bounds of legal pragmatism

When an orthodox application of the criteria for criminal responsibility does not meet the requirements of public policy, the law must change. When it is left to judges to make the change, existing institutions or concepts are likely to be adapted to meet social requirements.

In R v Gnango [2011] UKSC 59 (14 December 2011) Lord Kerr dissented in his orthodox application of the principles of party liability. He held that neither primary liability as principal offender nor secondary liability either as an aider, abettor, counsellor or procurer, or by reason of extended secondary liability (the sort of common enterprise-gone-wrong that in this case all judges agreed to call parasitic accessory liability) applied to the facts.

The facts were simple and are found in the statement of the question of law that arose in this case:

"If (1) D1 and D2 voluntarily engage in fighting each other, each intending to kill or cause grievous bodily harm to the other and each foreseeing that the other has the reciprocal intention, and if (2) D1 mistakenly kills V in the course of the fight, in what circumstances, if any, is D2 guilty of the offence of murdering V?"

It was a gunfight. There was no room for extended secondary liability here because any agreement that D2 may have had with D1 could not sensibly include agreement that D1 should shoot at him.

So D2 could only be guilty if he was a principal or if he aided, abetted, counselled or procured D1 in the killing of V who was an innocent by-stander. It should be obvious that he was not a principal, as he did not actually commit the murder himself. This was not obvious to Lords Brown and Clarke, and neither but to a lesser extent to Lords Phillips, Judge and Wilson. But they were engaged in extending the law for policy reasons.

The difficulty with orthodox secondary liability was that in this case the jury had not been invited to consider whether there was an agreement that D2 would be shot at, so even if this absurd possibility were a potential basis for liability it was not relevant to this appeal.

D2 could not have aided (etc) D1 in the killing of V unless he had helped (etc) by agreeing to be shot at.

Lord Kerr was correct in orthodox terms to conclude that there was no basis in the circumstances of this appeal to hold D2 liable for the murder of V.

That conclusion was not good enough for the other judges.

Lord Phillips and Lord Judge, with Lord Wilson agreeing, took the extremely pragmatic approach of saying it doesn't matter whether D2 was a principal or a secondary party, he and D1 both acted dangerously in a public place and each should be held accountable for V's death. Either could have killed someone and it was just fortuitous that the person who fired the fatal shot was D1. These judges, and Lord Dyson, preferred the secondary liability route to responsibility but they agreed with Lords Brown and Clarke that principal liability could also be used as the basis for liability.

Nor does the jury have to agree on the basis for liability: it is the conclusion as to guilt that requires agreement, not the route to that conclusion (63).

Well, you can't just pluck someone out of an unruly mob and say this person could easliy have been the one who caused the relevant harm so he should be held responsible for it even if it is known that he didn't actually do it himself. Nor can you pretend that he intentionally assisted or encouraged the commission of an offence when there is no evidence he meant to help or encourage its commission. Yes, the law must further the interests of the community, but there must be a rational, formalist, basis for attributing responsibility for crime. Otherwise we will have a society in which judges can simply say we shouldn't let this person off so we will hold him liable.

The strength of vagueness

Two vague but fundamental concepts

The first vague concept: abuse of process

Complicity by Australian officials in the unlawful deportation of the defendant (appellant) to Australia led to subsequent criminal proceedings against the defendant in Australia being stayed as an abuse of process in Moti v R [2011] HCA 50 (7 December 2011).

Abuse of process is open-ended, not to be confined to rigid categories of official misconduct (60):

" ... the forms of expression adopted in the decided cases must be understood in the context of the particular facts of each case. None should be read as attempting to chart the boundaries of abuse of process. None should be read as attempting to define exhaustively the circumstances of removal of an accused to this country that warrant exercise of the power to stay criminal proceedings against that person or as giving some exhaustive dictionary of words by one or more of which executive action must be described before proceedings should be stayed. None should be read as confining attention to whether any act of an Australian Government official constituted participation in criminal wrongdoing, whether as an aider and abettor or as someone knowingly concerned in the wrongdoing. And the use of words like "connivance", "collusion" and "participation" should not be permitted to confine attention in that way. All should be understood as proceeding from recognition of the basic proposition that the end of criminal prosecution does not justify the adoption of any and every means for securing the presence of the accused. And in this case, as in others, the focus of attention must fall upon what Australian officials did or did not do."

Recognition of abuse of process is a response to the policy of even-handed justice and the maintenance of public confidence in judicial process.

"57. ... two fundamental policy considerations affect abuse of process in criminal proceedings. First, "the public interest in the administration of justice requires that the court protect its ability to function as a court of law by ensuring that its processes are used fairly by State and citizen alike" [Williams v Spautz [1992] HCA 34; (1992) 174 CLR 509 at 520]. Second, "unless the court protects its ability so to function in that way, its failure will lead to an erosion of public confidence by reason of concern that the court's processes may lend themselves to oppression and injustice" [[1992] HCA 34; (1992) 174 CLR 509 at 520]. Public confidence in this context refers to the trust reposed constitutionally in the courts to protect the integrity and fairness of their processes. The concept of abuse of process extends to a use of the courts' processes in a way that is inconsistent with those fundamental requirements."

Here the official misconduct was in summary (63):

"...First, Australian officials (both in Honiara and in Canberra) knew that the senior representative of Australia in Honiara at the time (the Acting High Commissioner) was of opinion that the appellant's deportation was not lawful. Second, the Acting High Commissioner's opinion was obviously right. Third, despite the expression of this opinion, and its obviously being right, Australian officials facilitated the unlawful deportation of the appellant by supplying a travel document relating to him (and travel documents for those who would accompany him) at a time when it was known that the documents would be used to effect the unlawful deportation. That is, Australian officials supplied the relevant documents in time to be used, with knowledge that they would be used, to deport the appellant before the time for deporting him had arrived."

The majority, French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ, held that proceedings on the indictment should be permanently stayed.

Heydon J delivered an interesting dissent, focused on difficulties arising from the vague concepts concerning abuse of process and its lack of definition. Among the points he makes is the availability of alternative, disciplinary, responses to official misconduct instead of giving a person who may be guilty of serious offending immunity from conviction.

There are also some observations in this case on payment of prosecution witnesses, which was another ground of this appeal but which did not need to be considered in detail as no impropriety in that regard was held, unanimously, to have occurred.

The second vague concept: miscarriage of justice

In Handlen v R; Paddison v R [2011] HCA 51 (8 December 2011) the High Court held that the proviso could not be applied where a trial had proceeded on a mistaken appreciation of how participation in the offending could be proved. The requirements for secondary liability, namely that each appellant had intentionally aided, abetted, incited, counselled or procured the commission of the offence, should have been applied. (An alternative form of secondary liability was not relevant in this case.) But the trial proceeded wrongly on the basis that proof of membership of a joint criminal enterprise would be sufficient if commission of the relevant offence was part of that enterprise. The error is that not every member of such an enterprise is necessarily a party to every offence committed by members of the enterprise. This was overlooked by all counsel and by the trial judge. The Court of Appeal of the Supreme Court of Queensland had recognised the error but had applied the proviso because it was satisfied that the appellants were guilty, and there had not been a departure from the fundamental requirement of a trial according to law.

"47. As this Court explained in Weiss v The Queen, there is no single universally applicable description of what constitutes a "substantial miscarriage of justice" [[2005] HCA 81; (2005) 224 CLR 300 at 317]. The appellants were convicted of serious criminal offences ... following a trial at which the prosecution case was conducted, and left to the jury, on a basis for which the law did not provide. The conduct of the trial on this basis conferred an evidentiary advantage on the prosecution, leading to the admission of evidence to prove the existence and scope of the group exercise. Ultimately, the issue posed for the jury was whether the prosecution had proved that the appellants were parties to the group exercise when this was irrelevant to proof of their complicity in Reed's offences. The verdicts on the importation counts reflect the jury's satisfaction that each appellant was a party to the group exercise but it does not follow that the jury must have been satisfied of the facts necessary to establish the appellants' guilt of the importation offences in the only way for which the law allowed. It was not open to the Court of Appeal to apply the proviso in the circumstances of these appeals."

The majority ordered a new trial, Heydon J dissented and would have dismissed the appeals. He analysed the evidence and found guilt proved regardless of how the trial had been conducted. He did not see the defects as fundamental. For the majority, regardless of the strength of the evidence the trial had been such a departure from what was in accordance with the law that, in effect, the right to a fair trial was the dominant consideration.


Vague concepts can still be useful in the law. Reasonableness, fairness, interests of justice, the public interest, the weighing of values underlying rights, do not need to be defined as if they were mathematical concepts. Numbers too, when used in measurements, involve margins of error and require probabilistic reasoning. These areas of vagueness are opportunities for the exercise of judgment. Complaints about vagueness are like the formalists' complaints about pragmatism.

An unruly heckler at the back of the room might cry out that Heydon J had, in the first of these appeals, been too much the formalist, while in the second he had been too much the pragmatist.

Equality before the law, parity of sentence

Some general points on disparity of sentence can be gleaned from Green v R; Quinn v R [2011] HCA 49 (6 December 2011). The case needs to be read in its statutory context which includes a constraint on prosecution appeals against sentence. But aside from that,

• An appeal by an offender on the basis that his sentence was excessive by comparison to that imposed on a co-offender may be allowed to avoid disparity, even if the result is an inadequate sentence, as long as the result is not an affront to the administration of justice (French CJ, Crennan and Kiefel JJ at 33).
• A prosecution appeal against sentences imposed on some co-offenders should not give rise to a disparity with more lenient but un-appealed sentences imposed on other co-offenders (French CJ, Crennan and Kiefel JJ at 37).
• An appeal court should not introduce procedural unfairness by basing its judgment on a perceived defect in the decision of the court below when on the hearing of the appeal the defect was not mentioned and was not the subject of argument (French CJ, Crennan and Kiefel JJ at 76, 80).
• Intermediate courts of appeal should only overrule their own decisions infrequently and in exceptional circumstances, when the earlier decisions are manifestly wrong and not based on principle worked out in clear lines of useful authority (Heydon J dissenting in the result).
• Parity must not be conflated with proportionality, because the starting point is the appropriate sentence for the particular offence. Different offending may justify different sentences which should not be criticised as being disparate (Bell J dissenting, Heydon J agreeing, at 125).
  Prosecution appeals may not be allowed if the result would be disruption to rehabilitation, as the benefit of guidance to other courts may come at too high a cost in terms of justice to the individual (French CJ, Crennan and Kiefel JJ at 43).

Obviousness and obfuscation

Civil proof of offending

It should be obvious that proving something to the civil standard does not undermine failure to prove it to the criminal standard.

Blindingly obvious though this may be, it has troubled the Strasbourg judges. The UKSC was briefly troubled by it (R v Briggs-Price, noted here 2 May 2009), but more recently it was not: Gale v Serious Organised Crime Agency [2011] UKSC 49 (26 October 2011).

Again in Gale the context was civil proceedings for the recovery of property to a value equivalent to the proceeds of serious criminal offending. The judge had applied the civil standard of proof to the issue of the commission of such offences. The Supreme Court unanimously upheld that approach.

There was some delicate treatment of Briggs-Price, and what was the minority view there was now unanimously approved: 47-54. This was achieved by admirably obscure reasoning, perhaps inspired the confusing nature of the Strasbourg jurisprudence.

The presumption of innocence, and the finality of acquittals, are not compromised where in civil proceedings for the recovery of the proceeds of crime the court is satisfied on the balance of probabilities that criminal offending had occurred.



Waiver of the right to legal advice

Another thing that you would have thought was obvious is that a person can waive the right to legal advice without having to receive legal advice about waiver and its consequences.

This point had to be decided in McGowan (Procurator Fiscal, Edinburgh) v B (Scotland) [2011] UKSC 54 (23 November 2011).

Difficulties concerning the effectiveness of waiver can arise from the different rights that may be waived and from whether the alleged waiver was express or implied. But in all cases, waiver must be shown to have been "voluntary, informed and unequivocal" (Lord Hope at 17). "Informed" waiver will not exist if there is reason to believe the defendant had not understood the right that was being waived (36, 38, 44), but a defendant can understand without legal advice (46). The court will, however, proceed cautiously where the defendant is of low intelligence or is vulnerable for other reasons (47).

Further (49), if a defendant declines legal advice, the police should mention that such advice can be obtained by telephone, and if the defendant still declines, the police should ask why and record the answer. That, at least, would be best practice although not an absolute rule (50) [and see also Jude v HM Advocate (Scotland) [2011] UKSC 46 (23 November 2011)]. It could also be wise to advise the defendant that legal advice could be given without cost to him (51).

There was a time when courts took a more tight-sphinctered approach to what was adequate advice of the right to consult a lawyer. In R v Piper [1995] 3 NZLR 540 (CA) for example, it was thought to be unnecessary to inform the defendant that the consultation would occur without the police overhearing it. But now we have the Chief Justice's practice direction (issued under s 30 of the Evidence Act 2006) which requires privacy to be mentioned. It also requires mention of the availability of free advice.

Another point I should make about waiver is this. Waiver is a more sensible requirement to use than is consent. The issue should not be whether the defendant "consented" to police questioning, but rather whether he waived his right to legal advice and to silence. Similarly, in the context of searches, the issue should not be whether the defendant "consented" to a search, but whether he waived his right not to be searched. As anyone can see, it is incongruous to think that a defendant in possession of incriminating material would freely consent to being searched by the police, but less so to think that he might waive his right not to be searched.

Conviction appeals – burdens and risks

How sensitive should an appellate court be to the risk that an error at trial was sufficient to require the quashing of a conviction and the ordering of a retrial?

The proviso has been the traditional legislative criterion, and its formulation has usually used the concept of a substantial miscarriage of justice. That is, an error at trial would not require the quashing of a conviction if it was not a substantial miscarriage of justice.

There was therefore a distinction between miscarriages of justice that were not substantial, and those that were. The former could be termed harmless errors, and the latter errors that may have affected the result of the trial.

But what does "may have affected" mean for this latter group?

In R v Sarrazin 2011 SCC 54 the majority declined to draw distinctions between various grades of risk of affect on the outcome. At [27] Binnie J said:

"It seems to me that there is a significant difference between an error of law that can be confidently dismissed as "harmless", and an assessment that while the error is prejudicial, it is not (in the after-the-fact view of the appellate court), so prejudicial as to have affected the outcome. Such delicate assessments are foreign to the purpose of the curative proviso which is to avoid a retrial that would be superfluous and unnecessary but to set high the Crown's burden of establishing those prerequisites. The same can be said for the other branch of the curative proviso. As a result, the burden of the Crown to demonstrate an "overwhelming" case or a "harmless" error of law should not be relaxed."

What is interesting about this for those of us who are about to come under a reformed legislative regime is this. The burden has shifted from the Crown to the appellant on the issue of whether an error at trial was harmless. Formerly the Crown had to satisfy the appellate court that the error was harmless, now the appellant will have to satisfy the appellate court that it was not.

Only one submission to the select committee on the criminal procedure bill mentioned this point.

Section 232 of our Criminal Procedure Act 2011 materially avoids any distinction between substantial miscarriages of justice and those which are not substantial. The appeal court must allow an appeal if it is satisfied that a miscarriage of justice had occurred, and that means that there is a real risk that the outcome of the trial was affected by the error. There are other aspects, but these are the ones directly in point here.

How will an appellate court interpret "real risk"? Is there to be a gradation of risks, of the sort that the Supreme Court of Canada majority found unacceptable? Or is a real risk anything more than a harmless error or one which occurred in any case other than one where the prosecution's case was overwhelming?

As the dissent of Deschamps, Rothstein and Cromwell JJ in Sarrazin illustrates, opinions can differ over whether there was a reasonable possibility that an error could have affected the verdict.

In the circumstances of that appeal, the majority's approach, which was that failure by the judge to put to the jury the alternative of an attempt amounted to a substantial miscarriage of justice, is preferable to the minority's speculation on how the jury was thinking. Under the New Zealand reformed law, the majority's reasoning would lead to a conclusion that the trial had been unfair because it was not according to law. That is an alternative ground for finding a miscarriage of justice under s 232.

When our hair was black

In the course of considering the alleged naivety of Jonathan Sumption (UKSC blog 9 November 2011) I reached for my copy of "Equality" by Joseph and Sumption, which when it was hot off the press I reviewed for the University of Auckland students' newspaper in 1979.

Those were the days when Sumption's hair was black, as indeed was mine. Here is mine:

His was more flamboyant:

The book is an argument against taxation-imposed redistribution of wealth. The UK was then in the grip of rampant taxation of the rich. The authors' style has a formidable clarity that is only available to the very brilliant.

But although I now forget what I said about the book at the time (update: here is a copy of my review), it seems to me that insistence on logical rigor on matters of social policy has the same weakness I mentioned here on 3 November 2011. Relatively immature cognition favours logic over the weighing of values that should underlie policy reasoning.

Sumption has delayed taking his position on the UKSC bench so that he can work on an important case. He will receive a spectacular fee for that. In 1979 he and Joseph said (p 69):

"In practice, people will not work beyond the point at which the burden of the extra work exceeds the pleasure of the extra money."

Given that the work of a lawyer is generally not particularly burdensome, one can see how the balance may fall.

But people do work for reasons other than "the pleasure of the extra money." There are some burdensome areas of criminal law practice, as anyone who defends the children of the poor will realise. Legal aid returns can be insignificant compared to the burden of the work.

There is naivety in Joseph and Sumption's book. It seeks to shield its glib assumptions behind an insistence that only logic can rebut its argument. Dworkin's hedgehog has the better view. Compare the starkly different views of what is the business of the State. Joseph and Sumption (67):

"It is no business of the State to decide in advance what kind of society it ought to be governing and then to manipulate or frustrate known desires in such a way as to bring such a society into being. It must take its subjects as it finds them. It is means, not ends, which are the proper concerns of governments. It is their proper function to provide a framework of laws and institutions within which men can pursue ambitions of their own devising, and thereby create whatever society is the natural outcome of the infinite variety of human tastes and personalities."

Dworkin (352-354):

"Coercive government is legitimate only when it attempts to show equal concern for the fates of all those it governs and full respect for their personal responsibility for their own lives.

"... [E]verything the government of a large political community does – or does not do – affects the resources that each of its citizens has and the success he achieves. ... [T]he impact of ... personal variables on his actual resources and opportunities must in every case also depend on the political variables: on the laws and policies of the communities in which he lives or works.

"... [W]e cannot avoid the challenge of equal concern by arguing that the resources an individual has depend on his choices, not the government's choices. They depend on both."

Dworkin's conception of the purpose of government is more nuanced and more realistic.

Just to be even-handed, I should confess that around the same time as Joseph and Sumption were writing their short book I was writing an essay for a Masters tutorial on the hallmark requirement for the admissibility of similar fact evidence, in which I asserted that absence of a hallmark can itself be a hallmark. Even now my toes curl and I blush. True it may be that in some contexts the offender who has no hallmark may thereby be distinguished from other suspects, but more generally absence of a hallmark has no probative value. How young I was, how innocent. But the difference is that whereas my error was a simple slip of logic, easily corrected, Sumption's is a neglect of pragmatism. Experience tells us that weak judges tend to rely on logic at the expense of pragmatism. "The life of the law has not been logic, it has been experience."

Informer privilege

Both the state and a police informer may have an interest in keeping confidential the identity of the informer. But such confidentiality is subject to a defendant's right to a fair trial. A defendant may make his own inquiries into the identity of an informer, but this must be done with care to avoid conduct that obstructs the course of justice.

The line may be crossed, for example, if the defendant tries to obstruct prosecution proceedings by suggesting that he will disclose an informer's identity. In R v Barros
2011 SCC 51 (26 October 2011) that was alleged.

It is in the public interest that false claims of informer privilege are disallowed. The person may have been a participant in the offending to an extent that protection of confidentiality is not warranted. He may have instigated the offending. Both the defendant, and when the case gets to trial, the judge, have an interest in ensuring that claims to confidentiality and privilege are properly based.

Where there is a proper basis for informer privilege it is granted - without the balancing of competing interests that may be undertaken in other areas such as confidentiality relating to journalists' sources – but only if the defendant's right to a fair trial would not be compromised [33-35].

The Canadian approach seems consistent with that in New Zealand: Evidence Act 2006, ss 53, 64, 67.

NZSCBlog

[Update: the "NZSC blog" seems to have disappeared. I imagine that students are under too much pressure to produce quality beyond the limitations that their inexperience inevitably imposes. Not to worry, it was an ambitious plan.]

Congrats to the University of Auckland Law School on starting the New Zealand Supreme Court Blog. I enjoyed the analysis of Elias CJ's discussion of what academics call the third source of governmental power in Hamed v R.

I wonder if judges anonymously create their own blog sites and comment on each other's judgments. Perhaps they could adopt noms de keyboard and post rebuttals of critical assessments of their work.

I don't allow comments because my early experience was that they are just irritating, whether from judges or not.

Anyway, the new blog is off to a promising start, so I have added a link to it. This does not mean that I agree with its criticism of the CJ's judgment. It is possible that law schools still place emphasis on technical legal reasoning - identification of the ratio of a decision, recognition of precedents, distinguishing or applying other cases - which is formalist, at the expense of policy reasoning which is essentially pragmatic. The Supreme Court is a policy court. It is not bound by decisions of other courts, and is probably not even bound by its own decisions. It is concerned with finding the best solution to legal problems in the light of the judicially-perceived policy that best serves the current needs of our society. If it gets that wrong, the legislature can intervene.

Posner, in "How Judges Think" at 220-221 is particularly good on this topic, as are generally his chapters 7 and 8.

So, what sort of society do we want to live in? One where the police can approach people at random and ask if they may search their bags? People who think that the executive can do anything that is not specifically proscribed would have to accept that sort of society. Or do we prefer a society in which the police can only put questions to people when they have lawful authority to do so? I think our desire for freedom from executive interference supports this alternative. As it happened, the CJ did not support her conclusion with policy reasoning other than in a broad sense by reference to the significance of the Bill of Rights. Her formalist approach to what was essentially a question of pragmatism might reflect the schoolroom (but I prefer to doubt that), and has drawn formalist criticism. Both her approach and the criticism have mis-fired here.

I should add, while I am in this mood to be helpful, that a prime illustration of formalism being developed to an academic extreme and then discarded in favour of pragmatism is Field v R, discussed here on 27 October 2011.

What’s the word for ...?

Procedural fairness aims at facilitating substantive fairness. Sometimes procedural rights are so close to the right to a fair trial that there is no room to argue that notwithstanding breach of procedural fairness there was a fair trial. An argument along these lines was rejected in Abdula v R [2011] NZSC 130 (1 November 2011).

The Supreme Court in this case laid down procedural guidelines for the use of interpreters in trials in the future [60]. Those procedures had not been followed in this case. At [44] the Court said:

"The consequence of a breach of the right to the assistance of an interpreter under s 24(g) [of the New Zealand Bill of Rights Act 1990] is a breach of the right to a fair trial under s 25(a). We do not accept as correct the Crown's submission that, once a breach of the right to assistance of an interpreter is shown, the court must exercise a judgment as to whether the accused nevertheless had a fair trial. Rather, a properly established breach – the failure to meet the required standard – necessarily makes the trial unfair. In those circumstances, it is axiomatic that a substantial miscarriage of justice will have occurred. There can accordingly be no resort to the proviso under s 385(1) of the Crimes Act 1961.[Footnote: R v Condon [2006] NZSC 62, [2007] 1 NZLR 300 at [77]]"

In my view, this question of "failure to reach the required standard" marks the boundary between procedural unfairness and substantive unfairness. While the Court seems to be denying any distinction between the right to assistance of an interpreter and the right to a fair trial, it can only remove the distinction by employing the "required standard" concept. Really, the standard required of interpreters is that of avoiding trial unfairness.

Another interesting feature of this case is that the appellant has succeeded in establishing a rigorous procedure for future cases but not his own. The question for the Court then became, notwithstanding the departures in this case from the new standard was there any defect that resulted in an unfair trial here? The Court does here exactly what the Crown had "unsuccessfully" submitted it should do at [44] quoted above. But it does that by using the "required standard" analysis.

Failure of the appellant here was not surprising, because his counsel could not demonstrate any instance where failure of accurate interpretation impeded the conduct of the defence at trial [59]. Of course it would be difficult for counsel to do that here, as the words used by the interpreter were not known. On the facts, however, the defendant had at trial expressed no concern over the adequacy of the interpretation – but then, how could he know what was adequate?

In appeals like this (rape) there might be a public perception that the Court would do everything it could to avoid a retrial. Fortunately for the public perception of the administration of justice there seems to be sufficiently careful scrutiny of the circumstances in this judgment to put such concerns to rest.

Voices of reason

Thanks are due to UKSCBlog for drawing our attention to video interviews with some of the Justices, published by the Guardian.

We simply don’t do that

Acceptance of a bribe can occur after the performance of an act beneficial to the payer and without any expectation of payment when the act was done: Field v R [2011] NZSC 129 (27 October 2011).

This is because [59] "it is simply wrong for an official to accept money or like benefits in return for what has been done in an official capacity."

And [61] an environment would otherwise be created where

"(a) an official who receives such benefits will come to expect similar benefits in the future and is likely to act accordingly; and

(b) members of the public who know about, or suspect, what has happened will come to believe that unless they too provide such benefits, they will not receive dispassionate consideration and, if prepared to provide such benefits, will receive corresponding advantages."

Any expressions to the contrary as might be detected in previous cases or law commission reports didn't matter.

"[62] ... there is a fundamental inconsistency between the performance of official functions and the acceptance of private rewards for doing so. In large measure this is a corollary of the first reason in the paragraph above. But associated with this are related expectations about the way in which those in official positions, including Members of Parliament, can be expected to act. This consideration is also illustrated by the facts of the present case."

But there is room for courtesies:

"[65] ... if there is an exception, it must address the extent of the gift and the particular context in which it occurs. We consider, therefore, that there must be a de minimis defence in relation to gifts of token value which are just part of the usual courtesies of life."

As to the element that a bribe be accepted "corruptly",

"[66] ... In part it captures the requirement for a defendant to have acted knowingly. In the present case, this requirement required the Crown to establish that the appellant knew that the services he received were provided in connection with the immigration assistance he gave, meaning that he knowingly engaged in conduct which the legislature regards as corrupt. As well, it is the presence in s 103(1) (and like provisions) of the word "corruptly" which permits the de minimis exception to liability which we accept exists."

Deflecting the jury from its fundamental task

"The instructions which a trial judge gives to a jury must not, whether by way of legal direction or judicial comment on the facts, deflect the jury from its fundamental task of deciding whether the prosecution has proved the elements of the charged offence beyond reasonable doubt."

Hargraves v The Queen; Stoten v The Queen [2011] HCA 44 (26 October 2011) at [45] per French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ.

It was a potential breach of this for the judge to direct the jury that in assessing the defendant's evidence they could take into account his interest in the verdict.

Other examples of possible deflection of the jury from their task mentioned here are comment on the defendant's failure to give evidence [43] and asking the defendant in cross-examination why the complainant is lying [44].

Deflection of the jury from its fundamental task is a miscarriage of justice. Whether deflection has occurred in a particular case must be assessed in the context of the whole of the summing up and the circumstances of the case. As it turned out here, there was no miscarriage, because the jury would not have understood the judge to be saying that the evidence of a defendant must be scrutinised more carefully than that of other witnesses.

Omission? What omission?

For discussion of liability for omissions under the law of the Commonwealth of Australia, see Commonwealth Director of Public Prosecutions v Poniatowska [2011] HCA 43 (26 October 2011).

The majority held that legislation had failed to specify what act that was omitted was necessary for the actus reus of an offence of engaging in conduct that resulted in the obtaining a financial advantage that the defendant knew or believed he was not eligible to receive. Here, conduct is defined as including an omission to perform an act, and an omission is only a physical element of an offence if the law creating the offence makes it so.

The legislature, anticipating this glitch, introduced a new section (s 66A) into the relevant legislation. This enabled the requiring of a person to do a specific act in the relevant circumstances, namely to give information. Omission to do that specified act would apparently be sufficient for this element of the actus reus.

Heydon J, dissenting, criticised the increasing vagueness of statutory criminal law. He referred to

" ... the proposition advanced by the responsible Minister, when the Code was introduced into the House of Representatives in 1995, that it would reflect Benthamite ideals of certainty in the criminal law [footnote: Australia, House of Representatives, Parliamentary Debates (Hansard), 1 March 1995 at 1331]. One does not often encounter a more striking illustration of the vanity of human wishes. That is because very many parts of the Code, including the parts debated in this appeal, are inconsistent with those ideals. They represent a significant regression from the condition of Commonwealth, State and Territory criminal law as it was before 1995. That criminal litigation under the Code is conducted with any semblance of ordered justice is a tribute to the Australian legal profession, not to the Commonwealth legislature."

So, legislatures, sharpen your Act(s). Ho ho.

Some errors of law in judicial fact finding

For a reminder of the kinds of things that can be errors of law arising out of a trial judge's treatment of the evidence, see R v J.M.H. 2011 SCC 45. The headnote summarises them from the judgment of the Court delivered by Cromwell J at [24] – [32]:

"The Crown's right of appeal from an acquittal of an indictable offence is limited to 'any ground of appeal that involves a question of law alone'. The jurisprudence currently recognizes at least four types of cases in which alleged mishandling of the evidence may constitute an error of law alone giving rise to a Crown appeal of an acquittal; this may not be an exhaustive list. First, it is an error of law to make a finding of fact for which there is no evidence. However, a conclusion that the trier of fact has a reasonable doubt is not a finding of fact for the purposes of this rule. Second, the legal effect of findings of fact or of undisputed facts may raise a question of law. Third, an assessment of the evidence based on a wrong legal principle is an error of law. Fourth, the trial judge's failure to consider all of the evidence in relation to the ultimate issue of guilt or innocence is an error of law, but this error will be found to have been committed only if the reasons demonstrate that this was not done. The trial judge's reasonable doubt did not have to be based on the evidence; it could arise from the absence of evidence or a simple failure of the evidence to persuade him to the requisite level of beyond reasonable doubt. It is only where that reasonable doubt is tainted by a legal error that appellate intervention in an acquittal is permitted."

That decision (maple) tree again

If evidence was wrongfully obtained, what is the relevance to its admissibility that it could have been obtained lawfully?

Sometimes failure to use lawful means would aggravate the wrongfulness. Other times the lawful alternative would diminish the wrongfulness. When?

The Supreme Court of Canada got to grips with this in R v Côté 2011 SCC 46 (14 October 2011). The context for this sort of decision in Canada is R v Grant 2009 SCC 32, which I had some fun with here on 18 July 2009 for its complex model of how admissibility of improperly obtained evidence is to be decided. Grant was referred to recently by the Supreme Court of New Zealand in Hamed v R, discussed here on 19 September 2011.

So to the vital bit of Côté. The term "discoverability" means the ability to discover the evidence lawfully. Two "branches" of the Grant model have to be addressed: the seriousness of the misconduct, and its impact on the defendant Cromwell J delivering the joint judgment):

"[71] I turn to the first branch of the Grant test which is concerned with the seriousness of the Charter-infringing state conduct. If the police officers could have conducted the search legally but failed to turn their minds to obtaining a warrant or proceeded under the view that they could not have demonstrated to a judicial officer that they had reasonable and probable grounds, the seriousness of the state conduct is heightened. As in Buhay, a casual attitude towards, or a deliberate flouting of, Charter rights will generally aggravate the seriousness of the Charter-infringing state conduct. On the other hand, the facts that the police exhibited good faith and/or had a legitimate reason for not seeking prior judicial authorization of the search will likely lessen the seriousness of the Charter-infringing state conduct.

"[72] We come now to the effect of discoverability on the second branch of the Grant test — the impact on the Charter-protected interests of the accused. Section 8 of the Charter protects an individual's reasonable expectation of privacy. That reasonable expectation of privacy must take account of the fact that searches may occur when a judicial officer is satisfied that there are reasonable and probable grounds and authorizes the search before it is carried out. If the search could not have occurred legally, it is considerably more intrusive of the individual's reasonable expectation of privacy. On the other hand, the fact that the police could have demonstrated to a judicial officer that they had reasonable and probable grounds to believe that an offence had been committed and that there was evidence to be found at the place of the search will tend to lessen the impact of the illegal search on the accused's privacy and dignity interests protected by the Charter."

Nothing controversial there.

I still think the Canadian model in Grant is absurdly complex: who can visualise a 3-branch balancing exercise? The third branch is the public interest in admission of the evidence. It would be better to think of the first two as being one "branch", or arm of the balance, with a weight moving one way or the other along it according to the first two considerations.

In Grant the model was called a "decision tree". Draw it for us. I had a go, but couldn't say it would make comparison of cases easy.

Previous acquittals as propensity evidence

In Fenemor v R [2011] NZSC 127, the Supreme Court declined to establish a rule excluding, as propensity evidence, evidence of facts on which the defendant had previously been acquitted ("acquittal evidence"). The Court held that in each case admissibility of acquittal evidence will depend on its qualifying under s 40(1) of the Evidence Act 2006 as propensity evidence and then on its surviving the weighing of probative value against risk of unfair prejudice required by s 43.

The Court declined to give examples of when unfairness might make such evidence inadmissible, beyond that mentioned in Degnan [2000] NZCA 321, [2001] 1 NZLR 280 (CA) of an alibi having been the basis for the earlier acquittal. It was preferable that the case law should develop over time as questions of unfairness are very case-specific.

A challenge addressed to the acquittal dimension of the propensity evidence must demonstrate a logical connection between the acquittal and the claimed unfairness.In Fenemor it was argued unsuccessfully that the defendant would have to give evidence if the evidence was ruled admissible and that this was unfair. The Court, upholding the reasoning of the Court of Appeal, held that he would be in the same position if the evidence had led to a conviction, or if no earlier charge had been brought, so there was no logical connection between the acquittal and the claimed unfairness. Similar reasoning applied to a submission that the prior acquittal evidence would lead the jury to reason wrongly. Nor was it correct to argue that it would be unfair to adduce the similar facts that the previous jury had rejected, because that jury had only seen "one frame of what was now known to be a bigger picture" [22].

Assessment of the probative value of the evidence in relation to the present charge is different from assessing its value as proof of the earlier charge, and a priori there is no basis for a general conclusion that acquittal evidence has different probative value from propensity evidence that has not led to any charge [23].

Fruit of the poisoned tree

When may evidence obtained 'downstream' from improperly obtained evidence be admissible? By downstream I mean that the finding of the challenged evidence occurred as a consequence of the improper obtaining of the other evidence.

For example, if the defendant, under improper questioning, gives information to the police about a witness who could vouch for him, and the police consequently question that witness and obtain evidence incriminating the defendant, how should the court decide whether the incriminating evidence is admissible? Facts like this were considered in HM Advocate v P (Scotland) [2011] UKSC 44 (6 October 2011). This is a companion case to that considered in my note yesterday.

The position in domestic Scots law is that if the evidence could be led without reference to the improperly obtained evidence it would not be treated as inadmissible by reason of it having been found as a consequence of the improperly obtained evidence: Lord Hope at 17, citing Lawrie v Muir 1950 JC 19. Similarly, in England and Wales under s 76(4) and 78(1) of PACE 1978 all the circumstances must be considered and the consequential evidence is not invariably excluded: at 18.

As far as Strasbourg jurisprudence went, there was no definitive decision but guidance could be obtained from Gäfgen v Germany (2011) 52 ECHR 1 (noted here on 3 July 2008, Chamber, and Grand Chamber on 25 June 2010): Lord Hope at 22. No rule of automatic exclusion of such consequential evidence has been established 23.

The Supreme Court found some assistance from Canada: Thomson Newspapers Ltd v Canada (Director of Investigation and Research) [1990] 1 SCR 425, where the distinction was made between evidence created by impropriety, and evidence that may have gone undetected but for the impropriety but which nevertheless existed independently.

This was consistent with the conclusion Lord Brown reached 27:

" ... there is no absolute rule that the fruits of questioning of an accused without access to a lawyer must always be held to be a violation of his rights under article 6(1) and 6(3)(c) of the Convention. It is one thing if the impugned evidence was created by answers given in reply to such impermissible questioning. The leading of such evidence will be a breach of the accused's Convention rights unless there are compelling reasons to restrict the right of access: Cadder [v HM Advocate [2010] UKSC 43], para 55. It is another thing if the evidence existed independently of those answers, so that those answers do not have to be relied upon to show how it bears upon the question whether the accused is guilty of the offence with which he has been charged. So far as the accused's Convention rights are concerned, there is no rule that declares that evidence of that kind must always be held to be inadmissible. The question whether it should be admitted has to be tested, as in domestic law, by considering whether the accused's right to a fair trial would be violated by the leading of the evidence."

Lords Dyson, Kerr and Matthew Clarke agreed with Lord Hope, as did Lord Brown in a separate judgment. Lord Brown drew attention to how the admissibility balancing decision would be made in England and Wales under s 78(1) of PACE.

See also the attenuation approach in R v Wittwer (discussed here 6 June 2008), and occasions where a consequential link may be dispensed with but evidence nevertheless excluded: R v Ogertschnig (discussed here 26 October 2008). Improprieties may taint evidence without there being a consequential link, and as the present case shows, improprieties may be irrelevant notwithstanding such a link.

The right to legal advice

When does a person who is questioned by the police have the right to legal advice?

There was a tussle over this in Ambrose v Harris, Procurator Fiscal, Oban (Scotland) [2011] UKSC 43 (6 October 2011), and it turned on whether the UKSC should merely keep pace with Strasbourg jurisprudence or whether it should do more. By doing more it would extend a rule excluding evidence beyond the extent currently recognised. The majority were of the conservative view: Lord Hope at 17-20, Lord Brown agreeing 73, Lord Dyson agreeing 88, 105, Lord Matthew Clarke agreeing 116. Lord Kerr dissented on this point, referring to the Ullah-type reticence 126 of the majority and to the duties of courts to resolve issues 129 [see R v Special Adjudicator, ex parte Ullah [2004] UKHL 26]. He also mentioned 126 the "characteristically stylish twist" given to Lord Bingham's oft-cited dictum in Ullah in Al-Skeini [2007] UKHL 26 by Lord Brown (and see my note on the dissent in this case in the Grand Chamber [2011] ECtHR 1093).

So that's settled: the UKSC follows Strasbourg jurisprudence and does not extend it. The relevant Strasbourg rule excludes all confessional statements made under detention in police stations before the suspect had an opportunity to obtain legal advice. Here there were, in addition to the settled approach, policy reasons for not extending this rule: Lord Hope at 15, 58, 60; Lord Brown at 78, 87; Lord Dyson at 99-105; Lord Matthew Clarke at 120. Strictly, I shouldn't say the rule "excludes" the evidence, because Strasbourg leaves admissibility to national courts, but the finding of a breach of the fair trial right through use of the evidence will inevitably lead to a ruling of exclusion to avoid that unfairness, so it comes down to the same thing.

In Ambrose incriminating answers to police questions had been given in three cases at times before detention in police stations and when the defendants had not been given an opportunity to obtain legal advice.

The three cases concerned respectively (Ambrose) roadside questioning before breath alcohol procedures were undertaken, (M) questions at the defendant's home aimed at obtaining admissions to a serious assault, and (G) the questioning of a handcuffed defendant during execution of a search warrant at his flat. The Art 6(3) rights apply to a person "charged", and this has been held to mean when the person's situation was substantially affected, when suspicion was being seriously investigated and the prosecution case was being compiled (Lord Brown at 62).

The Court unanimously held that although none of the statements were inadmissible by virtue of a rule, admission of G's statements would be a breach of his Convention rights because he was detained in coercive circumstances within the concept of "charged" in Art 6(3). The majority held there was no breach of Art 6 in admission of Ambrose and M's statements but that the admissibility issue should be referred back for determination because there was arguably an issue of fairness.

Lord Kerr, dissenting on Ambrose and M, held that admission of all the confessional statements would be in breach of an extended application of Art 6 because the police believed each defendant had committed an offence and the answers would be given in evidence, and in also in relation to G, he was in custody.

The Supreme Court referred, in addition to Strasbourg jurisprudence, to cases from Canada (R v Grant [2009] SCC 32, noted here on 18 July 2009 – Lord Brown calling its judgments "immensely long" (81)) – and the USA (Miranda v Arizona 384 US 436 (1966)), and of course to its own decision in Cadder v HM Advocate [2010] UKSC 43 (noted here 27 October 2010) concerning persons detained in a police station.

Just as people in Scotland may feel a little irked at having the UKSC decide points of their criminal law, so too – we might reasonably suspect – may the UKSC feel irked at having the Strasbourg Court decide points of UK law. Lord Brown at 86: "...whatever else one may say about the Strasbourg jurisprudence ...". Humpf. And in furtherance of that attitude there was a marvelous sneakiness at play here: G got the evidence excluded and so received the benefit of what was effectively an extension of the very rule the Court majority was professing not to be allowed to extend.

No matter what you say ...

For a brief look at the requirements for establishing judicial bias, see Siemer v Heron [2011] NZSC 116. Orthodox response to a recusal application where the appellant relied on his own criticism of a judge as a basis for his claim of bias. Unsuccessful.

The NZSC on improperly obtained evidence

At last Hamed v R [2011] NZSC 101 is online, although in redacted form. It does not answer the question that will occur to most people: is it the duty of the police (1) to detect and stop offending, (2) to bring offenders to court with admissible evidence, or both of those, or some compromise between those?

The Prime Minister has indicated today that he will get legislation passed "suspending" the Supreme Court's decision in Hamed. Whatever that means, it is designed to ensure that other cases where covert surveillance has been used do not collapse because of exclusion of evidence obtained in that way. If in those cases, why not in this one too, since retrospective effect is intended? In any event permanent law reform is on the way "after the election".

This underlines the fact that it is not the police who are to blame for this (see my comment yesterday) but rather it is the fault of those who are responsible for giving effect to the Law Commission's recommendations.

But here I am concerned with the aspect of the case that will be of enduring interest to lawyers. This is the Supreme Court's first in-depth analysis of s 30 of the Evidence Act 2006.

The relevant question here is the extent, if any, to which the Supreme Court alters the interpretation of s 30 established in Williams v R [2007] NZCA 52, [2007] 3 NZLR 207 (CA) (not currently available online).

Where Williams is cited in Hamed, that is done with approval except on one point on which Tipping J departed from it at 240 by rejecting the Williams approach to assessing the seriousness of the offending. McGrath J also thought (277) that the maximum penalties were the guide although they were not a complete basis for assessment of the seriousness of the offending. But those were minority views, and Williams remains authoritative on s 30.

The application in Hamed of the balancing factors mentioned in s 30 needs to be considered. Broadly, it is orthodox. Aside from the minority difference just mentioned, there is some comment on the relevance of the centrality of the evidence to the prosecution case, but this only concerns whether this is relevant as a separate unspecified factor (Blanchard J at 201, Gault J agreeing at 281) or as part of the quality of the evidence factor in s 30(3)(c) (McGrath J at 276), but Tipping J dissented saying centrality is not relevant (236).

In applying the balancing criteria there was the usual variation in judicial opinion. The absence of an alternative investigatory technique was thought by Elias CJ to increase the seriousness of the impropriety (73), Blanchard J (196) appears to include this as a factor favouring admission of the evidence (Gault J agreeing at 281), Tipping J (246) also thought this pointed towards admission of the evidence, although not strongly, and McGrath J (274) treated this factor as increasing the reasonableness of the police misconduct. So, 4-1 absence of an alternative investigatory technique favoured admission.

The deliberate nature of the breach of the defendants' rights was an aggravating factor: Elias CJ 73, Blanchard J 194, Tipping J 233-234, but dissenting on this point McGrath J 267 thought that because the law was only now clarified by this case the deliberate nature of the breach did not enhance the gravity of the impropriety. And on the same point Gault J 284 did not agree with Blanchard J and held that the intrusion on the defendants' rights was no more serious that it was in relation to searches pursuant to warrant. Again, but by a different 3-2, the deliberate nature of the breach of rights did weigh in favour of exclusion.

The Supreme Court of Canada got some favourable mention, particularly for some of its dicta in Grant (see my discussion of that case here on 19 July 2009, and here twice on 18 July 2009). Tipping J at 230 approved dicta on the desirability of taking the long view of the repute to the justice system as opposed to responding to the immediate public clamour for conviction. Blanchard J 187 noted that the decision is not just a balancing of impropriety against the need to convict, but rather it comes down to the need to avoid bringing the system of justice into disrepute. Elias CJ 58 made the same observation, after saying 57 "It would be wrong to treat the assessment of proportionality as being the same in all cases of impropriety" – by which I think she means the division between exclusion and admission of the evidence is not marked by a straight line (see my discussion of improperly obtained evidence).

So, all in all (carefully working through the judgments to find the majority on each point) this is an orthodox application of s 30, even when the seriousness of the offending is assessed at an enhanced level because of public safety considerations.

Search, surveillance and the Urewera case

A newspaper report today has a commentator suggesting that confidence in the police will be reduced because of their handling of the investigation of what has come to be known as the Urewera terrorist case.

Until the Supreme Court decided the issue in a judgment not yet available on the usual web sites, R v Hamed [2011] NZSC 101, and which I have not yet seen, the law on whether search warrants could authorise surveillance was unclear. Differences among judges in this case reflected that lack of clarity.

As the New Zealand law Commission noted in its report "Search and Surveillance Powers" NZLC R97 2007, there are few references to surveillance powers in the statutes (see para 11.19 of the report). None of those are relevant to the Urewera case. The NZLC recommended that legislation should be formulated to clarify surveillance powers.

In the absence of a statutory or regulatory framework the courts have had to consider whether surveillance is a kind of search, and this has turned on the circumstances of each case. It was held to be not a search in a case which may still be subject to name suppression so I just give its neutral citation: [2010] NZCA 294, and also in another such case: [2010] NZCA 287. Sometimes surveillance from a neighbouring property by consent of the neighbour has been held to be lawful: R v Beri (2003) 20 CRNZ 170 (CA), and R v Robertson [2009] NZCA 154. And sometimes surveillance has been held to be a search: R v Gardiner (1997) 15 CRNZ 131 (CA). In other cases the courts have held that it was unnecessary to decide whether surveillance is a search because the issue of the admissibility of the evidence turned on the balancing exercise used at common law and which is now enacted in s 30 of the Evidence Act 2006: R v Fraser [1997] 2 NZLR 442, R v Peita (1999) 17 CRNZ 407.

The actions of an undercover officer approaching the defendant's door and secretly making a video recording of the ensuing events has not been held to be a search: R v Smith (Malcolm) [2000] 3 NZLR 656 (CA), although in the possibly suppressed case mentioned above, [2010] NZCA 287 the Court of Appeal noted that opinions may differ over whether this was a search, so it determined the admissibility of the evidence on reasonableness and balancing grounds.

Where the police obtained a search warrant and installed video surveillance equipment, the Crown conceded that that was illegal, however the Court of Appeal doubted whether that concession was correct: (another possible name suppression case) [2010] NZCA 457.

So even right up to the Supreme Court's decision in the Urewera case the law was unclear as to whether surveillance was necessarily a search. The point was if surveillance was a search, it could be authorised by a search warrant. If it was not a search and could not be legitimised in that way, and if it involved a trespass the admissibility of evidence obtained by such surveillance was governed by the balancing exercise in s 30. In the Urewera case, the High Court held that surveillance could not be authorised by a search warrant, but the Court of Appeal overruled that decision. The Supreme Court overruled the Court of Appeal on this point. It also overruled the Court of Appeal and the High Court on the admissibility of some of the evidence where it related to less serious charges.

[Update: as it turned out, the Supreme Court's analysis of search was more subtle than this. It differentiated between a narrow form of search that could be authorised by a warrant - search for things that existed at the time the warrant was executed, where "things" did not include captured images - and the wider form of search that was addressed in s 21 of the Bill of Rights. Covert surveillance is a search when it infringes reasonable expectations of privacy, but it cannot be authorised by a search warrant. There could be an unreasonable search without a trespass: Blanchard J at 57, 63, 64 of [2011] NZSC 101, another aspect of which is discussed here tomorrow.] 
 
My impression from media reports as the case has made its rather slow way through the courts is that the police acted in good faith in their approach to collecting the evidence. I don't see any reason for the public to lose confidence in the police over this aspect of the case.

A small collection ...

No compensation for judicial breach of rights

If a judge breaches your rights under the Bill of Rights, you may not claim compensation: Attorney-General v Chapman [2011] NZSC 110 (16 September 2011) per McGrath and William Young JJ jointly with Gault J concurring. Elias CJ and Anderson J dissented.

The majority reasoning emphasised public policy which was an extension to the personal immunity of judges from suit and was based on the need to protect judicial independence.

Eyewitness identification evidence

An illustration of circumstances where there was insufficient support for the reliability of identification by an eyewitness who claimed in a fleeting sighting to recognise the defendant from two encounters several years previously (there being no evidence about why those should have made the present identification more reliable), so that the requirements of s 45 of the Evidence Act 2006 were not met, is Harney v Police [2011] NZSC 107 (16 september 2011). The Court emphasised the need for strict compliance with s 45 in view of the dangers of mis-identification. Dock identifications too should be permitted only in the most exceptional circumstances [20 at footnote 20]. The witness's confidence in the accuracy of his own identification is just a factor to be taken into account, as the opposite side of the coin of hesitancy, and confidence cannot itself satisfy a reliability test [33].

Mootness and ordering judge alone trials

Occasionally an appellate court may hear and decide an appeal notwithstanding that the issue in question no longer is a live one in the case: Signer v R [2011] NZSC 109 (16 September 2011), applying R v Gordon-Smith [2008] NZSC 56 (its substantive point discussed here on 23 March 2009) and see also R v McNeil [2009] SCC 3 discussed here on 20 January 2009).

The substantive point in Signer concerned the interpretation of s 361D of the Crimes Act 1961. Does the word "likelihood" in subsection 3(b) mean that a judge must find it "probable" that jurors will not be able to perform their duties effectively? Or does it just mean that the judge considers there is an "appreciable risk" of that? The Supreme Court held that "likelihood" imports a balancing exercise and does not set down a specific standard. The accused's right to a jury trial may be outweighed by the need for a fair trial. Beyond saying that, the Court preferred not to explore the matter without a live issue which would provide specific circumstances for consideration.

More on Momcilovic

I should, so as not to appear lazy, add a few observations on the aspects of R v Momcilovic (last entry) concerned with interpretation.

It is not unusual to find in Bills of Rights provisions to the effect that rights shall only be subject to such limitations as are justified in a free and democratic society. Section 32(1) of the Victorian Charter of Rights and Freedoms is an example. So too is s 5 of the New Zealand Bill of Rights Act 1990. The question is, are these provisions to be used in defining what the rights mean?

In Momcilovic the minority (French CJ, Crennan and Kiefel JJ) answer was no, s 32(1) is just a mechanism by which the court can decide whether to issue a declaration that the legislation in question inconsistent with the rights. However the majority (and I recall Dworkin's comments on the weights that might be given to individual judge's votes: Justice for Hedgehogs, pp 484-485) did use the rights-limiting provision as interpretive: Gummow J at 166-168, Hayne J agreeing at 280, Heydon J at 411-427, and Bell J at 678. Similarly, but where the New Zealand BORA does not give the courts power to make declarations, s 5 has been held to be interpretive, to be used to determine what the right means: Hansen v R [2007] NZSC 7 per Blanchard, Tipping, McGrath and Anderson JJ (57-60, 89-92, 186, 190-192), but with Elias CJ dissenting at 6, 7, 15-24. The Chief Justice's dissent takes the approach which was favoured by the minority of the High Court of Australia in Momcilovic.

But even this interpretive role in the New Zealand approach is not necessarily applied. For example, in Morse v Police [2011] NZSC 45, discussed here on 6 May 2011, only McGrath J used it.

The interpretive method, outlined by McGrath J in Hansen at 192, applies the following steps:

1. Ask first whether the circumstances fit within the ordinary meaning of the statutory provision being applied. Here, which standard of proof would be a natural interpretation of the reverse onus provision?
   
2. Then ask whether, on that meaning there appears to be an inconsistency with a protected right. Here, is the legal burden on the defendant inconsistent with the right to be presumed innocent?
   
3. If there is such an inconsistency, ask whether the limit on the right is a justifiable one in terms of s 5. Here, is the legal burden on the defendant justifiable?
   
4. If the limit is not justifiable, ask whether there is another meaning available through which the statute can be read consistently with the right. Here, can the reverse onus provision be read in a way that puts an evidential burden on the defendant?
   
5. If there is no such other meaning, the natural meaning must be applied. Here, the legal burden on the defendant.
   

For the fourth step, which is interpretive, the powers given by the relevant interpretive provision must be considered. Is the court limited to applying the ordinary and natural meaning of the legislation, or can the court apply a meaning that "can be given" to the enactment, or thirdly, is the court allowed to be more imaginative and strain the words of the statute to preserve the relevant right? In Momcilovic French CJ, Crennan and Kiefel JJ applied the ordinary rules of statutory construction, including the principle of legality, to determine the meaning of the reverse onus provision. This was the first of these interpretive approaches, although these judges were denying that the rights limiting procedure was relevant and were in effect just using step 1. Accordingly, no straining for an unnatural but rights-consistent meaning was permitted. There was no asking what might be reasonable limits on the right to be presumed innocent. But the minority agreed with the majority that the construction contended for by the defendant was not available on ordinary principles of statutory interpretation: to "satisfy" the court means more than merely to raise a reasonable doubt.

It is well worth reading French CJ's judgment in Momcilovic at 37-51 for a comparison with the approach in applying s 3 of the Human Rights Act 1998[UK], in particular at 49:

" ... Section 3 of the HRA has a history and operates in a constitutional setting which is materially different from that which exists in Australia. Before its enactment, United Kingdom courts, which had to give effect to the supremacy of European Community law, lacked domestic legislation providing for the direct application of rights under the ECHR. In the result there was a perception that British judges were denied the responsibility of safeguarding Convention rights and that the European Court of Human Rights had become "in effect a supreme constitutional court of the UK."[Footnote 97: "Lester, Pannick and Herberg (eds), Human Rights Law and Practice, 3rd ed (2009) at 12 [1.34]."] The HRA was enacted under the political rubric of "bringing rights home"[Footnote 98: "Lester, Pannick and Herberg (eds), Human Rights Law and Practice, 3rd ed (2009) at 12-15 [1.35]-[1.46]."]."

Perhaps this is a clue for anyone who, like me, is perplexed about why s 3 of HRA should be given a different meaning to s 6 BORA, as it was in Hansen and Lambert, as was mentioned in yesterday's note. The UK enactment applies the provisions of a Convention, whereas the NZ one is technically just an ordinary statute. In the absence of some sort of rationale like this, one is left with the thought that there is a bit of judicial pride being protected.


The UK interpretive approach is of the third kind mentioned above: some straining for a rights-compliant meaning is allowed (Ghaidan v Godin-Mendoza [2004] UKHL 30 noted here 19 September 2005), the New Zealand and Australian approaches are of the second kind.

And for a wee caustic-to-the-point-of-being-truthful glimpse of how the law works, see Heydon J in Momcilovic at 455.

Reverse onus provisions and the presumption of innocence

A reverse onus provision is one that puts a burden of proof on the defendant. It raises questions about what standard of proof is required to meet that burden. Illustrations commonly occur in legislation concerning drug offending.

For example, s 6(6) of the Misuse of Drugs Act 1975[NZ] applies when a person has been proved to have been in possession of a specified quantity of a drug, and it operates by creating a presumption that the person had the purpose of supplying it, "until the contrary is proved". This has been held to mean that the defendant has the legal burden – that is, to the standard of the balance of probabilities – of proving absence of that purpose: Hansen v R [2007] NZSC 7. The Supreme Court held that "proved" means proved on the balance of probabilities, and it rejected an interpretation of "proved" which was that it means the burden of raising a reasonable doubt, that is, the evidential burden. For my earlier commentary, see 13 January 2010, 5 March 2009, 20 February 2007, and 19 September 2005.

This latter, rejected, interpretation of "proved" was based on a suggestion by Glanville Williams in "The Logic of 'Exceptions'" [1998] CLJ 261, 265. I had relied on that article in making submissions (also rejected) to the New Zealand Court of Appeal in R v Phillips [1991] 3 NZLR 175. In that case, Cooke P for the Court found that interpretation of "proved" a "strained and unnatural interpretation".

The House of Lords did not find it "strained and unnatural" in R v Lambert [2001] UKHL 37, and indeed at [42] Lord Steyn said:

"In Kebilene [ [2002] 2 AC 326 ] I described this as a respectable argument: 370G. Lord Slynn agreed: 362A. Lord Cooke of Thorndon regarded the distinguished author's view as a possible meaning under section 3. Specifically, Lord Cooke stated that "unless the contrary is proved" can be taken to mean "unless sufficient evidence is given to the contrary" [2000] 2 AC at 373G. I respectfully adopt Lord Cooke's observation. Applying section 3 I would therefore read section 28 (2) and (3) as creating an evidential burden only. In particular this involves reading the words "prove" and "proves" as meaning giving sufficient evidence."

So, something about s 3(1) of the Human Rights Act 1998[UK] allowed Lord Cooke (who was the same person as Cooke P in Phillips) to accept the Glanville Williams interpretation of "proved", whereas the corresponding provision of the New Zealand Bill of Rights Act 1990, s 6, had allowed neither him nor the Court in Hansen to accept it.

Much can turn on the interpretive directions given by the constitutional legislation.

The same submissions about the reverse onus were made in Momcilovic v R [2011] HCA 34 (8 September 2011). The wording of the reverse onus provision here was different and was held to be unambiguous, in that it requires the defendant to "satisfy the court" of the specified matter. "Satisfy" has always been understood to mean to establish on the balance of probabilities. See French CJ at 62, Crennan and Kiefel JJ jointly at 581 and Bell J at 659. That applied to proof of possession of the drug. But as to possession for the purpose of trafficking, the legislation did not impose a presumption and so the possession and the purpose had to be proved by the prosecution to the standard of beyond reasonable doubt: French CJ at 72, Gummow J at 200 (Hayne J agreeing at 280), Crennan and Kiefel JJ at 611 and Bell J at 659.

Momcilovic is, in addition, full of dicta of interest to those who are concerned with the Australian constitutional legislation.

For the curious ...

To begin the eighth year of this site, I reflect on the rewards and pleasures of scholarship, here.

Scientific research on eyewitness identification evidence

Every criminal law practitioner and judge should read the amicus brief filed by the American Psychological Association in Perry v New Hampshire (cert granted, No 10-8974). It refers to research on the risks of errors in eyewitness identification evidence, and on the utility of judicial warnings to juries.

(Thanks again to Peter Tillers of Cardozo School of Law at Yeshiva University for pointing this out.)

Now we are seven

That's seven years of this blog!

Usually I mark the anniversary by some repulsively self-indulgent boasting, smugly superior conceit or obnoxiously vain self-citation.

Not this time.