Discuss: no case or no chance?

When should a judge prevent a prosecution case from continuing and discharge the accused for lack of evidence?

The "no case" submission was addressed in DPP v Varlack [2008] UKPC 56 (1 December 2008). This was a case in which the prosecution had relied on inferences, rather than on direct evidence of involvement in a murder.

There is potential in this subject for confusion of a number of threads. One thread is what the judge must decide in ruling on a no case submission. Another is how the jury must be directed on inferences. Another is how an appellate court decides whether an error may have deprived the accused of a real chance of a more favourable verdict.

In Varlack the Privy Council endorsed a statement of the law by King CJ in Questions of Law Reserved on Acquittal (No 2 of 1993) (1993) 61 SASR 1, 5, which included the following:

"It is not [the judge's] concern that any verdict of guilty might be set aside by the Court of Criminal Appeal as unsafe. Neither is it any part of his function to decide whether any possible hypotheses consistent with innocence are reasonably open on the evidence ... ."

So, never mind how the jury will be directed about inferences of innocence, and never mind about how an appellate court will view a conviction. It is beside the point on a no case submission that the judge will tell the jury that they cannot convict if they think there is a reasonable inference that prevents an essential element of the prosecution case from being established to the necessary standard.

But it is relevant on a no case submission that the judge considers that the evidence on an element of the offence is so inherently weak or vague that a reasonable jury properly directed could not convict. It is only in that class of case, where the evidence is tenuous, that the judge's assessment of the acceptability of the prosecution evidence is relevant to the no case issue.

That is because questions of fact are for the tribunal of fact. On a no case submission, if the evidence is not inherently weak or vague, the judge has to decide whether the evidence is adequate. Adequacy is a question of law. It does not require anticipating what the jury will decide. A case is adequate, in this sense of overcoming a no case submission if, to quote the "canonical statement of the law" by Lord Lane CJ in R v Galbraith [1981] 1 WLR 1039, 1042 cited in Varlack at 21:

"... a reasonable jury properly directed could on that evidence find the charge in question proved beyond reasonable doubt."

King CJ elaborated, in the passage quoted by the Privy Council at 22 of Varlack:

"I would re-state the principles, in summary form, as follows. If there is direct evidence which is capable of proving the charge, there is a case to answer no matter how weak or tenuous the judge might consider such evidence to be. If the case depends upon circumstantial evidence, and that evidence, if accepted, is capable of producing in a reasonable mind a conclusion of guilt beyond reasonable doubt and thus is capable of causing a reasonable mind to exclude any competing hypotheses as unreasonable, there is a case to answer. There is no case to answer only if the evidence is not capable in law of supporting a conviction. In a circumstantial case that implies that even if all the evidence for the prosecution were accepted and all inferences most favourable to the prosecution which are reasonably open were drawn, a reasonable mind could not reach a conclusion of guilt beyond reasonable doubt, or to put it another way, could not exclude all hypotheses consistent with innocence, as not reasonably open on the evidence."

So, as the Board put it at para 23, the test is what inferences a reasonable jury properly directed might draw, as distinct from what inferences the judge (or the Court of Appeal) thought could or could not be drawn.

You may think that here there are moments of clarity, rapidly followed by periods of obscurity. The point to remember is that the facts are for the jury to determine except in cases where the evidence is inherently weak or vague.

The position was summarised in R v Harder 29/5/03, Williams J, HC Auckland T002481:

"A conviction will be unsafe within the terms of the authorities only rarely and only where the 'necessary minimum evidence to establish the facts of the crime has not been called' so that the elements have not been proved or in the rare case where the evidence is plainly so inadequate and threadbare or the witnesses so discredited that the trial Judge concludes that a conviction must necessarily be unsafe and the interests of justice therefore compel discharge."

I think the position would be clarified by this sort of omission of reference to the jury. The issues are whether there is evidence on each element of the case, and whether any of that evidence is plainly so inadequate and threadbare or the witnesses so discredited that the trial Judge concludes that a conviction must necessarily be unsafe and the interests of justice therefore compel discharge.

In Varlack the Board held that the trial judge had correctly rejected the no case submission, and the respondent's conviction was restored. On another point, it held that a good character direction should have been given even though the accused's defence involved admissions of bad conduct, but that here the omission of that direction had made insufficient difference to the trial to affect the verdict.

For the sake of a bit of mental exercise, one might compare the law on no case submissions to that on applications of the proviso. The latter, insofar as relevant here, involves considering whether, in the absence of an error that occurred at trial, the jury could have acquitted the accused. So whereas the no case issue is whether the jury could convict, the proviso issue is whether the jury could acquit.

As can be seen from the cases on the proviso (see Index to this site, entries under "Proviso" and "substantial miscarriage of justice"), the courts are not of one mind on how to decide whether a jury could acquit. Again, the difficulties arise from the court having to decide what a jury could do.

Just for the sake of tossing around ideas, what if the appellate court's task were to be seen as deciding whether there is a defence case that should go to a jury? The error at trial would have to link to the defence case, but, assuming it did, should the court quash a conviction and order a retrial if there is a defence case that is not tenuous (inherently weak or vague)? Each side should meet the same criteria of adequacy on the question of whether its case should go to the jury, whether before or at trial (prosecution case considered by the judge) or by way of retrial (defence case considered on appeal). Discuss.

Breach of right to legal advice

Failure by the police to advise a suspect – now the accused - of his right to legal advice can result in exclusion of evidence against him. This exclusion might also extend to evidence of things found as a consequence of what he said. That, while exciting, is not the thing that interests jurists. They are interested in the basis for exclusion of that evidence.

Is the basis for exclusion the need to ensure a fair trial, or is it the need to uphold broader policies such as the need to maintain control over the conduct of agents of the state and to prevent the administration of justice being brought into disrepute?

Why does the difference between these bases for exclusion matter? Because a different form of decision process is appropriate to each. On the one hand, there are decisions that proceed by way of application of rules. In these, if the right has been breached, exclusion of the tainted evidence is an automatic consequence. On the other hand, there are decisions that involve a balancing of competing interests. In these, the seriousness of a breach of the right is weighed against the public interest in holding the accused responsible for the alleged wrongdoing, and the balance is determined by applying a criterion of proportionality: whether exclusion of the tainted evidence would be a proportionate response by the court to the seriousness of the breach in this case.

The right to a fair trial is an absolute right, and it seems inapt to apply to it a decision process that involves balancing of competing interests. So it is necessary to know whether breach of the right gives rise to fair trial concerns, or to wider public policy concerns.

Cases noted here in previous blogs illustrate the different approaches. See the Index
, and in particular Carr v Western Australia (HCA,blogged 25 October 2007), R v Singh (SCC, 2 November 2007), and Gafgen v Germany (ECtHR, 3 July 2008).

In Salduz v Turkey [2008] ECtHR 1542 (27 November 2008) the Grand Chamber addressed the consequences of a breach of this right (para 55):

"...the Court finds that in order for the right to a fair trial to remain sufficiently "practical and effective" ... Article 6 § 1 requires that, as a rule, access to a lawyer should be provided as from the first interrogation of a suspect by the police, unless it is demonstrated in the light of the particular circumstances of each case that there are compelling reasons to restrict this right. Even where compelling reasons may exceptionally justify denial of access to a lawyer, such restriction - whatever its justification - must not unduly prejudice the rights of the accused under Article 6 ... . The rights of the defence will in principle be irretrievably prejudiced when incriminating statements made during police interrogation without access to a lawyer are used for a conviction."

This indicates some qualification on the need to give the suspect the information about his right to legal advice ("... as a rule ... unless ... compelling reasons to restrict this right"), and if the circumstances of the case are such that the right has been breached this means that there would be "in principle" irretrievable prejudice if the evidence is admitted. This comes close to a rule of exclusion. It would be a rule if the Court had said "The rights of the defence will, as a matter of principle be irretrievably prejudiced ...".

Supporting the view that the Court was not laying down a rule are the points made when the Court applies the "principles" to the present case (para 56 - 63). These are rather confusing as they include comments on matters the Court does not have to decide (58) and they refer to the applicant's youthfulness while also noting that the police procedure was being applied to everyone (60, 61). The Court focuses on how the breach of rights affected the applicant's trial (57, 58), in particular noting that the trial court treated the tainted statement as central evidence which other evidence in the case confirmed. This approach does not look like the application of a rule of exclusion.

If the decision process is not the application of a rule, it is not necessary to refer to breach of the right as going to the fairness of the trial. Policy fairness, not trial fairness, should be the context for the decision. However the Court, without defining trial fairness, considered that this breach did go to trial fairness.

I do not suggest that trial fairness is necessarily the wrong basis for the admissibility decision. A fair trial is one where the law is accurately applied to facts determined impartially. It could be argued that absence of legal advice might give rise to an unreliable statement, so that if the court relied on it that would skew the proceedings by causing partiality (giving improper weight to prosecution evidence). It could even be argued that such a statement would not be voluntary in the sense that the law requires, and that the use of such evidence would be of questionable reliability and would similarly render the trial partial.

The trouble with that view, focused as it is on the proper weight of the tainted evidence, is that questions of the weight of the evidence are usually for the tribunal of fact once the evidence is admissible. A coerced confession is not necessarily unreliable. Its rejection by the court is really a matter of public policy.

Courts in New Zealand tend to justify exclusion of evidence tainted by breach of the right to legal advice on unreliability grounds (see the cases mentioned in Fig 4 of my draft paper on improperly obtained evidence). Nevertheless, as that paper shows, the pattern of decisions follows that of decisions based on balancing of competing values. It seems to be a matter of the right results for the wrong reasons.

Issue estoppel: rule or discretion?

What place do rules have in modern evidence law?

Should the admissibility of evidence of prior misconduct, for which the accused has been tried and acquitted, be determined by a rule, such as issue estoppel, or by a discretion?

That was the main issue in R v Mahalingan [2008] SCC 63 (14 November 2008).

Before looking at how this was addressed in that case, it is worth recalling that the rules of evidence should promote the ascertainment of the truth, that a trial must be fair to the accused but fairness to the prosecution is also important although not overriding, and that evidence of an alleged fact does not have to establish that fact beyond reasonable doubt before it is admissible as part of the case for either side.

This last point is illustrated by Bayes' Theorem.

In Mahalingan the Supreme Court unanimously ordered a retrial on aggravated assault. After the original trial on that charge the accused had been acquitted of attempting to pervert the course of justice, it having been alleged that he had telephoned a witness before the aggravated assault trial and tried to persuade that person not to incriminate him. Should evidence of that alleged telephone conversation be admissible at the retrial on the aggravated assault?

The minority (Deschamps, Abella and Charron JJ) approached this consistently with the propositions about the law of evidence that I have set out above. To be admissible, the evidence of the telephone call need only be established on the balance of probabilities (166).

That standard is applied as a matter of policy, and is not a requirement of logic, as Bayes' Theorem demonstrates.

The minority rejected an argument that the rule concerning issue estoppel should apply in criminal law. Various reasons for this rejection were advanced by Charron J, who delivered the joint judgment of the dissenters. Most importantly she held that (132):

"The application of issue estoppel as an absolute rule precluding the admission of prior acquittal evidence in all cases regardless of context is inconsistent with the basic rule favouring the admission of all relevant evidence, subject to countervailing concerns."

The countervailing concerns are adequately expressed in existing concepts and provisions: abuse of process, character evidence admissibility, rules against multiple convictions and collateral attack, and the pleas of autrefois acquit and autrefois convict; s 11(h) of the Charter is also relevant (84). Such of those matters as were relevant would be considered by the judge at the retrial, and the evidence was not rendered inadmissible by issue estoppel which is only an appropriate rule for civil cases.

McLachlin CJ delivered the judgment of herself and Binnie, LeBel, Fish and Rothstein JJ. Issue estoppel is retained in criminal law but in a revised form (to correct its drift into difficulty) so that it now precludes the prosecution from relitigating an issue that has been resolved in the accused's favour in a previous criminal proceeding, whether on the basis of a positive fact finding or on a reasonable doubt (2, 22, 26).

The policies behind this are the desirability of avoiding inconsistent verdicts, the need to uphold the principle of finality, and the need to ensure fairness to the accused.

This last matter, fairness to the accused, does not mean fairness in the sense of a fair trial, but rather it means fairness in the public policy sense. The Chief Justice elaborated her use of the expression "fairness to the accused" at 39:

"In my view, it is clear that fairness to the accused requires that an accused should not be called upon to answer allegations of law or fact already resolved in his or her favour by a judicial determination on the merits. This is the most compelling rationale for retaining issue estoppel in criminal law, as it goes to the core tenets of our criminal justice system. The state has the right to charge an accused and to prove the facts at a trial of the charge. If a judge or jury conclusively decides a fact in favour of the accused, including via a finding of a reasonable doubt on an issue, then the accused should not be required in a subsequent proceeding to answer the same allegation. To require, in effect, a second defence of the issue would be to violate the fundamental function of res judicata."

This requires a determination, in each case, of whether the issue in question has been previously decided in the accused's favour (33):

"Only issues that were either necessarily resolved in favour of the accused as part of the acquittal, or on which findings were made (even if on the basis of reasonable doubt) are estopped."

That, I suggest, will operate capriciously. The accused has the burden of showing that the issue was decided in his favour (23), and his ability to meet this burden may depend on what he was previously charged with, and how many issues he raised at the earlier trial. He might diminish his chances of successfully relying on issue estoppel if he raised multiple issues, even though a genuinely innocent accused would put everything in issue.

In New Zealand we have crept away from issue estoppel in criminal law: R v Davis [1982] 1 NZLR 584 (CA), Bryant v Collector of Customs [1984] 1 NZLR 280 (CA), R v Coombs [1985] 1 NZLR 318 (CA), R v Fatu [1989] 3 NZLR 419 (CA), R v Degnan [2001] 1 NZLR 280 (CA), Fox v A-G [2002] 3 NZLR 62 (CA). See also Z v Dental Complaints Assessment Committee [2008] NZSC 55 (blogged here 25 July 2008) especially at paras 57-59, 125-127, for discussion of the use of criminally determined matters in civil proceedings, a topic also addressed in abuse of process terms, without reference to issue estoppel. That trend suggests that the minority approach in Mahalingan would be favoured here. We have regarded the jurisdiction to prevent an abuse of process by, inter alia, excluding evidence on public policy grounds, as a potentially powerful means of ensuring fairness in the wider sense. This contrasts with the majority judgment in Mahalingan, where abuse of process was described as a broad, somewhat vague concept that varies with the eye of the beholder, reserved for obviously egregious abuses of the Crown power, to be relied on successfully only extremely rarely (42):

"To protect parties from relitigation, abuse of process would need to be cast in a less discretionary form than it now takes. Therefore, considering the high threshold for proof and the unpredictability of its operation, it is unlikely that the doctrine of abuse of process adequately achieves the fairness goal that underlies the doctrine of issue estoppel."

That may reflect the emaciated form of the abuse of process jurisdiction in Canada. In other legal systems it may be relatively (I feel I am entitled to use the word "relatively" in view of my recent brush with smugness) strong and an adequate substitute for the rule against issue estoppel.

I should add, lest I be accused of regarding Canadian law as unduly flaccid, that the Charter provides the remedy of discretionary exclusion (s 24(2)) in terms that will usually make resort to the common law unnecessary: R v O’Connor (1995) 103 CCC(3d) 1 (SCC) per L’Heureux-Dube J at para 70-72. Under s 24(2) the determination is whether admission of evidence obtained in a manner that infringed or denied a Charter right or freedom would bring the administration of justice into disrepute. Common themes had been evident between the common law and Charter jurisprudence. For example, in R v Mack (1990) 44 CCC(3d) 513 (SCC) Lamer J, for the Court, addressed police conduct of an investigation that amounted to entrapment, and held that admitting evidence obtained through such conduct would bring the administration of justice into disrepute. He noted (para 77) that the same principle applied with respect to the common law doctrine of abuse of process. The dominance of the Charter does not render irrelevant Lamer J’s statement of basic principles in Mack.

Nevertheless, the majority in Mahalingan seem to distrust the usefulness of abuse of process. True it is that criticisms for vagueness can always be levelled at concepts that can only be expressed in general terms. Harmless fun can be had in making such criticisms, as occurred in argument in Ngan v R SC8/2007, 14 August 2007 at pp 32-33 (jibes which misfire, as the “high sounding principles” in the Charter are also in the New Zealand Bill of Rights Act 1990). And Blanchard J, the jester in that argument, set out his own general principles in R v Shaheed [2002] 2 NZLR 377 (CA) in defining the criterion for discretionary exclusion of improperly obtained evidence, a formulation that has been adopted in s 30(2)(b) Evidence Act 2006[NZ]: “...determine whether or not the exclusion of the evidence is proportionate to the impropriety by means of a balancing process that gives appropriate weight to the impropriety but also takes proper account of the need for an effective and credible system of justice.” [emphasis added; cf para 156 of Shaheed]

Judges do have a sense of what such general concepts mean. As cases applying them accumulate, patterns emerge and these aid the prediction of how the general concepts will work in particular instances.

Exclusion of improperly obtained evidence

Users of my method for analysing discretionary exclusion of improperly obtained evidence will be interested to see that in R v Horsfall [2008] NZCA 449 (30 October 2008) the Court ruled inadmissible evidence obtained through an improperly conducted search where the offending had attracted a sentence of over 4 years imprisonment. I have included this in my draft paper.

I could do it in my sleep

Does it matter that the judge sleeps during significant parts of a trial? What if there is a jury, can the judge sleep longer? What if the judge is alert for most of the time, and only misses some bits of evidence, through sleep or conscious inattention?

In Cesan v R [2008] HCA 52 (6 November 2008) the High Court of Australia considered two appeals against convictions that had resulted at a trial by jury where the judge's sleeping had distracted the jury. French CJ summarised the facts (94):

"In this case the facts ... indicated a substantial failure of the judicial process. The trial judge was asleep on a number of occasions on the 11 days when evidence was being given. He slept at least once on most of those days and on some days on two or three occasions. Some of the sleep episodes, possibly between two and five, lasted from 10 to 15 minutes. Most lasted between two and 10 minutes. The judge's sleeping was accompanied by heavy breathing on a number of occasions and he snored when Cesan was giving his evidence. This was disruptive and caused Cesan to look around at the trial judge. The judge also slept through parts of [the other accused] Mas Rivadavia's evidence. The jury was distracted by the judge's sleep episodes and some of the jury members found his behaviour amusing and even emulated it."

It was sufficient for French CJ that there was an appearance of injustice. This made the likelihood of convictions irrelevant (97). The flaw in the proceedings had been fundamental and incurable. It was a miscarriage of justice by failure of the judicial process, constituted by the judge's substantial failure to maintain the necessary supervision and control of the trial (96). Therefore the proviso could not be applied.

That straightforward reasoning, which separates procedural fairness from the question of the likelihood of the convictions being justified on the evidence, was not adopted by the other members of the Court.

A joint judgment was delivered by Hayne, Crennan and Kiefel JJ, with whom Gummow J and Heydon J agreed. These reasons stress (104, 110) that established principles are being relied on. This, in effect, means that the obscure approach laid down in Weiss v R (blogged here 16 January 2006) is going to make this difficult.

And it does. The joint judgment reasons that because the jury were distracted it is not possible to conclude from the written record of the trial that guilt was proved beyond reasonable doubt (127). That non sequitur, repeated at 130, emerges from an attempt to apply the Weiss approach of perceiving the likelihood of guilt as it appears from the record of evidence but also taking into account the verdicts.

It is difficult to see why the verdicts should be of assistance to an appellate court in the task of reviewing the adequacy of the evidence, given that the jury were misled in some way – usually by a misdirection on the law, or by inadmissible evidence, or as here by judicial slumber.

It would have been preferable for Gummow J, instead of simply agreeing (107) with the joint judgment's approach to the proviso, to have taken the line he took in AK v Western Australia (blogged here 27 March 2008), reasoning that this miscarriage was substantial and there was no need to ask if the accuseds were guilty. The absence of reasons in the judge alone trial in AK is not all that far removed from the absence of juror attention to the evidence here.

In Cesan the majority took one approach to the proviso, namely that it cannot be applied where the court cannot conclude that the convictions were supported by the evidence (127). The other generally accepted reason for not applying it – procedural unfairness at trial regardless of the likelihood of guilt – provides a more straightforward reason, as French CJ demonstrated, for ordering a retrial of each appellant.

Overkill

By laying too many charges the prosecution may give the accused a get-out-of-jail-free card.

This happened in R v JF [2008] SCC 60 (31 October 2008).

There was one death in this case, but the accused faced two charges of manslaughter. The jury convicted on the more serious, and acquitted on the lesser; this led the SCC to quash the conviction because the verdicts were inconsistent. The accused was left with two acquittals.

The accused's foster child had died as a result of injuries sustained at the hands of the accused's wife, who pleaded guilty to manslaughter. The two charges brought against the accused were

1. Manslaughter by criminal negligence;
   
2. Manslaughter by failure to provide the necessaries of life.
   

Deschamps J was the sole dissenting judge. She was able to perceive a distinction between the elements of the charges that provided a basis for the different verdicts. This distinction is the kind of thing that (in my view) you can persuade yourself you can see if you don't look directly at it, but as soon as you do it seems insubstantial. The jury may have had a reasonable doubt on the second charge, she reasoned, because they did not accept the Crown had proved an element peculiar to it, namely that the accused's failure had the effect of endangering V's life or of causing permanent endangerment to his health.

Deschamps J's judgment illustrates how a complicated explanation is unlikely to be convincing. Look at paras 73 and 74. Given that the jury concluded that the Crown had not proved that the accused's failure to provide the necessaries of life had endangered V's life, or caused or likely caused his health to be endangered permanently, how could the jury then conclude that on manslaughter by criminal negligence his marked and substantial departure from what a reasonable parent would do in the circumstances caused V's death?

On the first charge, the jury found (and correctly so in Deschamps J's opinion) that the accused should have been alert to the risk to V's safety.

Fish J delivered the judgment of himself, McLachlin CJ, Binnie, Abella, Charron and Rothstein JJ. The verdicts were inconsistent because each charge required proof that the accused had failed in his duty to protect V. The real distinction between the charges was, as Fish J said at 11, that in the present context the criminal negligence charge requires a marked and substantial departure from the conduct of a reasonably prudent parent, whereas the other charge simply requires a marked departure.

The second charge had the lesser degree of fault, and acquittal on it meant that the jury could not have found proved the higher degree of fault in the first charge. The jury should have considered the first charge only if it found the second proved.

Juries may well be more pragmatic that the law permits. There is a niggling possibility here that the jury simply thought that there should be one conviction, which should be for the more serious charge.

The Crown should have put its case to the jury in the way indicated by Fish J: first consider the charge of manslaughter by failure to provide the necessaries of life. If the verdict on that is guilty, then consider the charge of manslaughter by criminal negligence. If the verdict on the failure to provide the necessaries charge is not guilty, then the verdict on the criminal negligence charge must also be not guilty. Since it did not do this, the Crown lost on both.

The law appears to be different in New Zealand, where the charges would both require the same "major departure" from the standard of care expected of a reasonable person: s 150A, 160(2)(a) and (b) Crimes Act 1961 and R v Powell [2002] 1 NZLR 666 (CA), so they could not be put in the way indicated in R v JF. In such a case it would be necessary to make clear what the difference between the charges was alleged to be, otherwise only one could be left to the jury.

Reducing convictions

If juror ignorance about the meaning of "beyond reasonable doubt" had the effect of causing too many acquittals, you can be sure something would be done to improve judicial directions on the point.

The New South Wales Bureau of Crime Statistics and Research, in Bulletin No 119 (September 2008) "Juror Understanding of Judicial Instructions in Criminal Trials" by Lily Trimboli, concludes on this topic:

"The problems in juror understanding ... do not all stem from the way judges give instructions. It is assumed at common law that the phrase 'beyond reasonable doubt' requires no explanation and is readily understood by most ordinary people. Appeal courts have, for this reason, repeatedly warned trial court judges (see Green v The Queen (1971) 126 CLR 28, 32-33) not to attempt to clarify the phrase when explaining its importance to juries. The present study shows, however, that there is considerable divergence among jurors about the meaning of 'beyond reasonable doubt'. Around half (55.4%) of the jurors surveyed, believed that the phrase 'beyond reasonable doubt' means 'sure [that] the person is guilty'; 22.9 per cent believed that the phrase means 'almost sure' the person is guilty; 11.6 per cent believed that it meant 'very likely' the person is guilty; and 10.1 per cent believed it meant 'pretty likely' the person is guilty. This is quite a wide spread of opinion and it suggests that some clarification of the threshold for convicting a person would be of assistance to juries. While statutory clarification of the phrase 'beyond reasonable doubt' may be controversial, one relatively simple change that could be made to improve current practice would be to provide written materials to the jury to assist in their deliberations."

A perfectly sensible suggestion, but what should a written instruction say?

In discussing R v Wanhalla (25 August 2006) I noted the Court of Appeal's reluctance to mathematise the concept of proof beyond reasonable doubt by expressing it as a probability of guilt. I suppose that people vary in their understanding of "sure", "almost sure" and "very likely". The survey does not seem to have attempted to establish a way of ranking those words against a common reference, other than the phrase beyond reasonable doubt itself. If the jurors in the survey had been asked to express "sure", "almost sure", "very likely" and "pretty likely" as probabilities a recommendation could have been made as to the usefulness of directing juries in terms of probabilities.

The question in the survey was put as "In your view, does the phrase 'beyond reasonable doubt' mean pretty likely the person is guilty/very likely the person is guilty/almost sure the person is guilty/sure the person is guilty?"

What is clear is that there are likely to be too many convictions because of juror misunderstanding of the standard of proof. Progress in rectifying this seems to be painfully slow; such dragging of the feet by officialdom (I use this awful word to avoid using another awful word) would not occur if there were too many acquittals.

See also on this, Jeremy Gans ....

Over-zealous handcuffing

One of the limitations of only looking at the decisions of the senior appellate courts is that sometimes first instance decisions that perfectly illustrate significant principles are missed.

Thanks to this excellent Canadian blog our attention is directed to this decision of the Provincial Court of Alberta: R v Ogertschnig[2008] ABPC 293 (16 October 2008).

On a charge of refusing to supply a sample of breath, the judge found (14):

"I am satisfied on balance that the accused had been fully cooperative with Officer Renaud and had unhesitatingly agreed to provide a roadside sample. I find that when he exited his vehicle to do so, he was thereupon immediately handcuffed by Officer Renaud and pushed towards the police vehicle, approximately fifty feet away, and shoved into the back while handcuffed. During the process of placing him in the back of the police vehicle, the brim of the accused's hat was pushed down low onto his forehead, thereby impairing his vision other than in a downward direction."

Three independent reasons supported dismissal of this charge: insufficient grounds for requiring the breath sample, reasonable excuse for refusal to provide the sample arising from the improper use of handcuffs, and – of particular interest here – a breach of s 7 of the Charter, concerning liberty and security of the person, which constituted grounds for excluding the evidence of the defendant's refusal to supply the sample of breath.

Here there was a causal and temporal connection between the unlawful arrest and the refusal to supply the sample. It is clear that if, on different facts, there had been no causal connection but only a temporal one, the evidence could still have been excluded: R v Wittwer (blogged here 6 June 2008). An illustration would be where the improperly handcuffed driver continued his cooperation with the procedures, as occurred in the New Zealand case Police v Chadwick [2002] DCR 880. The handcuffing did not cause the evidence to be obtained, but there was a temporal and contextual connection.

I have discussed Wittwer and Chadwick in Misuse of Drugs at para 1300, as follows:

"The exclusion of evidence that was obtained improperly is now governed by s 30 Evidence Act 2006. Other provisions of the Act are concerned with exclusion of unreliable statements: s 28, and exclusion of statements influenced by oppression: s 29. Those provisions operate when there is a causal connection between impropriety, or a source of unreliability, or a source of influence of oppression, and the obtaining of the challenged evidence. Outside of such causal nexus, the public policy discretion is likely to remain: s 11 Evidence Act 2006. Occasions where there is a nexus, other than of a causal nature, are where there is a temporal or contextual association between an impropriety and the obtaining of the evidence, as illustrated in R v Wittwer ... . In Police v Chadwick [2002] DCR 880 the charge of driving with excess blood alcohol was dismissed because of improper use of handcuffs after the defendant had failed a breath test, but before the blood sample was taken; here the impropriety did not cause the obtaining of the blood test result, as the defendant was "co-operating to the full" (para 15), but it was sufficiently serious – illegal, unreasonable, and in breach of s 23(5) New Zealand Bill of Rights Act 1990 - for the evidence to be excluded and the charge to be dismissed on policy grounds. This conclusion was reached upon consideration of Shaheed[[2002] 2 NZLR 377; (2002) 19 CRNZ 165]... . If Chadwick had been decided after the commencement of the Evidence Act 2006 it would still be a case of contextual impropriety, not within the scope of the Act, where the public policy discretion involved a balancing exercise that is essentially the same as that which is provided for in s 30."

Ogertschnig illustrates how lawful resistance to improperly imposed restraint can result in exclusion of evidence, and Chadwick illustrates the common law exclusion of evidence tainted by impropriety, a remedy that should continue to exist notwithstanding the enactment of the discretion to exclude evidence where there is a causal connection between the impropriety and the obtaining of the evidence.

On mixed statements

The Supreme Court of Canada has simplified the direction that judges are to give juries on mixed statements. These are statements by an accused person that are partly incriminatory and partly exculpatory. Now, juries are not to be told that the exculpatory parts might be given less weight because the accused was not under oath when they were stated or because they were not subject to cross-examination: R v Rojas [2008] SCC 56 (24 October 2008).

This is a rejection of the English approach to directing juries on mixed statements, which was established in R. v. Duncan (1981) 73 Cr. App. R. 359 and applied in R. v. Aziz [1995] 2 Cr. App. R. 478. Under this approach, it is proper for the judge to tell the jury that exculpatory parts of a mixed statement do not necessarily carry the same persuasive weight as incriminatory parts.

The cases suggest that an approach like that in Rojas has been approved in New Zealand: R v Tomkins [1981] 2 NZLR 170 (CA), R v Tozer [2002] 1 NZLR 193, (2001) 19 CRNZ 269 (CA), and – to pick an Australian example cited in Tozer – in Western Australia: Middleton v R (1998) 19 WAR 179 (CCA WA). Those cases support the proposition that it would be wrong to tell a jury that the exculpatory parts of a mixed statement may have less weight because they were not given under oath, the jury should be told to consider the whole of the statement and to give such weight to such portions of it as it thinks proper.

It has, however, been by no means unusual to hear judges suggest to juries that there are various circumstances that indicate that the exculpatory parts of mixed statements should be given little weight, while adding that these are matters for the jurors to decide for themselves. Indeed, judges could hardly sum up prosecution cases without referring to such circumstances. But the point is that under Rojas the weight differential is not a matter of law.

The wider context of all the circumstances of the case led the Court to conclude, in Rojas, that the misdirection was "more unfortunate than fatal" (41). This was particularly so because the judge had explained that a denial may raise a reasonable doubt as to guilt, that an accused was entitled to the benefit of the doubt, and the judge had left the assessment of the statements entirely to the jury.

On the other hand, in a case decided the same day, the misdirection on mixed statements was fatal: R v Illes [2008] SCC 57 (24 October 2008). In contrast to Rojas, this was a majority decision. Here the jury may have thought that it was a matter of law that incriminating parts of mixed statements carry more weight than do exculpatory parts. The proviso could not be applied because the jury deliberations had been lengthy, the judge had given the misdirection after the jury had indicated they were at an impasse, and the mixed statements were critical evidence. The majority concluded that the jury would not inevitably have reached the same verdict if they had been properly directed.

Plainly, where the mixed statements do not form a critical part of the prosecution case, there will be significant evidence to suggest that their exculpatory parts deserve little weight; in such circumstances it would, as Illes illustrates, be harmless error for the jury to think that this reduced weight applied as a matter of law.

The common law rules as to how an accused's statements to the police should be treated in evidence developed as an exception to the rule against hearsay. As an exception, their contents were admissible as proof of their truth, and this was, over time, recognised as applying to the exculpatory parts as well as to the incriminatory parts. In New Zealand the relevant law has become statutory: Evidence Act 2006. This has not necessarily made the position simpler.

Section 27 provides that the hearsay rules do not apply to evidence offered by the prosecution in a criminal proceeding of a statement made by a defendant. It is not clear that the prosecution offers the exculpatory parts of a mixed statement, although the argument would be that these were offered as evidence of the defendant's lies. Section 124 provides for lies directions, but it will not always be the case that the defence would want this sort of direction given. Section 21 provides that if a defendant in a criminal proceeding does not give evidence he may not offer his own hearsay statement in evidence. Are the exculpatory parts of his mixed statement hearsay? Not if they are offered by the prosecution (s 27). If the prosecution omits reference to them, the defence may elicit them in cross-examination, and this amounts to "offering" them as evidence (s 4 definition of "offer evidence"), but because of s 21 the defence may only do this if the defendant is going to give evidence. Yet the defence may not have decided at that point whether the defendant will give evidence. Could the exculpatory parts of a mixed statement be admissible as ordinary hearsay evidence, under s 18, as s 18 is not subject to s 21? The problem with that is that there is a high reliability threshold for the admission of hearsay evidence, namely that the circumstances relating to the statement provide reasonable assurance that the statement is reliable: s 18(1)(a). Exculpatory statements are hardly likely to carry that assurance, otherwise the defendant would not be on trial.

I suspect that "a statement" in s 27 means the whole statement that was made, not just the incriminatory parts. Section 6(c) requires fairness, and the admissibility of exculpatory parts of mixed statements arose at common law out of considerations of fairness. Section 21 should be read restrictively, so that the exculpatory parts of the non-hearsay statement are offered not by the defendant but by the prosecution. There is no provision in the Act specifically addressing the nature of an appropriate direction that the jury should be given about the weight of exculpatory parts, but s 122 does permit the topic to be addressed and no particular form of words is required. The question whether to follow the Supreme Court of Canada's approach in Rojas will need to be decided, but it could be argued that s 122 pre-empts the courts' power to develop particular rules.

Fantasy and invention

In discussing recent invention (see blog for 8 July 2008 concerning R v Barlien, and the further comment on 19 July 2008), I suggested that an "invention" could be an innocent invention and not just a dishonest one. The point has been addressed briefly in R v Stewart [2008] NZCA 429 (22 October 2008).

Here there were several themes discernable in the cross-examination of the complainant. One, that in the Court's view did not necessarily amount to an allegation that she had recently invented her evidence, was the suggestion that she had been infatuated with the accused and had fantasised about these purely imaginary events. The Court did not refer to another possible difficulty with using an allegation of fantasising as the basis for admitting a prior consistent statement, namely the likelihood that such a prior statement would also be the product of inflamed imagination.

The Court's indication that, on its own, an allegation that a complaint was the result of fantasising need not be an allegation of recent invention, seems to suggest that any innocent misstatement of the facts would not count as an "invention" for the purposes of s 35(2) of the Evidence Act 2008. Personally, I don't see why not, as that provision refers both to use of prior consistent statements to respond to challenges to the witness's veracity and to the witness's accuracy. Inaccuracy can occur without untruthfulness.

The infatuation suggestion in Stewart had, however, to be seen in the context of the other themes to the cross-examination in this case. These were allegations of deliberate falsehood and motivation to obtain financial advantage. In the particular context the Court held that it was not possible to separate the allegations into those which were said by the defence to be motivated only by fantasy from those that were said to be motivated by dishonesty. Accordingly, the defence position was that the complaints were recent inventions, with the consequence that, subject to s 8 of the Evidence Act 2006 (the general discretion to exclude unfairly prejudicial evidence), evidence of the complainant's prior consistent statements was admissible.

The Court referred at 85 to R v Barlien and followed it on the point that the prior consistent statements were evidence of their truth.

Our fair-minded feminine side

Barristers, being fair-minded observers of the judiciary, are constantly amazed at how fair-minded judges are. (I smirk, but then I often smirk – it doesn't mean anything.) No doubt judges have acquired this fair-mindedness in the course of their practice at the bar before they were appointed – elevated can hardly be the right word – to the bench. No doubt barristers acquire their own fair-mindedness by learning from the example of judges. Aside from the chicken and egg question that comes to mind, we are justified in rejoicing at this perpetual cycle of fair-mindedness.

Occasionally the complaints of clients make it necessary for counsel to submit, reluctantly but courageously, that a judge has been biased, and so an examination of the concept of fair-mindedness is required. The House of Lords, back from their surprisingly long summer break, have addressed this in Helow v Secretary of State for the Home Department [2008] UKHL 62 (22 October 2008).

They don't say so, but in describing the qualities of fair-mindedness the Law Lords are portraying themselves. There is, as modesty demands, a gentle tone of mockery in Lord Hope's opening remarks:

"The fair-minded and informed observer is a relative newcomer among the select group of personalities who inhabit our legal village and are available to be called upon when a problem arises that needs to be solved objectively. Like the reasonable man whose attributes have been explored so often in the context of the law of negligence, the fair-minded observer is a creature of fiction. Gender-neutral (as this is a case where the complainer and the person complained about are both women, I shall avoid using the word "he"), she has attributes which many of us might struggle to attain to."

Too modest, and a signal we are in for some sarcasm on the sexual-politics front. So,

"2. The observer who is fair-minded is the sort of person who always reserves judgment on every point until she has seen and fully understood both sides of the argument. She is not unduly sensitive or suspicious, as Kirby J observed in Johnson v Johnson (2000) 201 CLR 488, 509, para 53 [I interpose here to note that Kirby J in that passage is absurdly conscientious in avoiding use of the male pronoun]. Her approach must not be confused with that of the person who has brought the complaint. The "real possibility" test ensures that there is this measure of detachment. The assumptions that the complainer makes are not to be attributed to the observer unless they can be justified objectively. But she is not complacent either. She knows that fairness requires that a judge must be, and must be seen to be, unbiased. She knows that judges, like anybody else, have their weaknesses. She will not shrink from the conclusion, if it can be justified objectively, that things that they have said or done or associations that they have formed may make it difficult for them to judge the case before them impartially."

A further quality of this creature of fiction is that he is "informed":

"3. Then there is the attribute that the observer is "informed". It makes the point that, before she takes a balanced approach to any information she is given, she will take the trouble to inform herself on all matters that are relevant. She is the sort of person who takes the trouble to read the text of an article as well as the headlines. She is able to put whatever she has read or seen into its overall social, political or geographical context. She is fair-minded, so she will appreciate that the context forms an important part of the material which she must consider before passing judgment."

Some caution is needed when a judge has been associated with a political cause, as Lord Walker said (26), adding that the judicial oath is not a panacea (27). Lord Cullen noted that some faith could be placed in the judge's training and experience (30) – just as I thought. Lord Mance referred briefly to a principle of automatic disqualification (40) that could arise in circumstances that didn't apply here. He also observed that cogent evidence is needed to show judicial bias (57) but he also described the judicial oath as a symbol, not a guarantee, of impartiality. As to whether a judge's failure to disclose an interest is necessarily an indication of lack of fair-mindedness, things get a bit circular when Lord Mance points out (58) that failure to disclose might indicate real fair-mindedness because the judge was so fair minded that it wouldn't have occurred to him that he might not be fair-minded.

Well it's all marvellous stuff, and very reassuring.

Too much information

Anyone who thinks criminal cases are really just civil cases but with a higher standard of proof as to the ultimate issue will not be surprised by the growth in complexity of pre-trial procedures. Those relating to disclosure of information by the prosecution to the defence received some consideration by the Privy Council in McDonald v HM Advocate [2008] UKPC 46 (16 October 2008).

The need to ensure that an accused person receives a fair trial seems to be taken to mean that, as far as is possible, the "parties" (to use an expression from the civil side of the law) should be put on an equal footing. The assumption is that the accused will have to prove his innocence, regardless of what might be said about the standard of proof.

Fraying of the golden thread has led to development of a new "golden rule", stated by Lord Bingham in R v H and C [2004] 2 AC 134 and referred to by Lord Rodger in McDonald at para 50:

"...Put shortly, the Crown must disclose any statement or other material of which it is aware and which either materially weakens the Crown case or materially strengthens the defence case...".

The application of this rule requires development as cases arise. Reference was made in McDonald to the current preparation of a legislated code concerning disclosure in Scotland.

In New Zealand we have new legislation that has not yet come into effect: the Criminal Disclosure Act 2008. This will impose obligations that are much wider than those of the golden rule. [Update: the Act commenced on 29 June 2009.]

In McDonald Lord Rodger was particularly concerned with the problems that will arise when large volumes of information come under consideration for possible disclosure. He pointed to the different functions of the prosecution and the defence in an adversarial system, 60, and observed that there is no way of avoiding reliance on officials carrying out their duties conscientiously. He considered that where the defence makes only a general request for disclosure it would need to be able to satisfy a court that the material it sought would have a legitimate bearing on an issue before disclosure would be ordered (76).

The narrow rule/wide request/narrow enforceability approach is not taken in the New Zealand legislation, which has more of a wide rule/wide request/fairly wide enforceability structure.

Recently there was some public concern over the extent to which the defence could obtain knowledge about people who were not involved in a particular case although their personal information had been gained in the course of police inquiries. That concern is heightened where accused persons represent themselves and have access to that sort of information. Perhaps that incident involved an over-eagerness to be candid with the defence, although it could also indicate an unwillingness by the police to devote time to the sifting of information to isolate that which might assist the defence. The Act is limited to "relevant" information (s 13(2)(a)), and this would not authorise the police to simply dump as much information as possible on the defence.

Our ethos is different from yours

In a decision that very nearly passed me by – because of its civil nature – the Supreme Court of Canada has held that the civil standard of proof is fixed, does not change according to the seriousness of the issue or the improbability of the allegation, and a flexible approach to it is not in law appropriate : FH v McDougall [2008] SCC 53 (2 October 2008).

This last point contrasts with the New Zealand Supreme Court's majority decision in Z v Dental Complaints Assessment Committee (blogged 25 July 2008). A fundamental difference in ethos is apparent: the SCC does not hesitate in this unanimous judgment to acknowledge that a change in the law as it has hitherto been applied is appropriate. The NZSC majority are more conservative, declaring that a flexible approach to the civil standard of proof has worked up to now, so there is no reason to change it. So although the standard in NZ is fixed, it is - strange to say - applied flexibly.

It will be recalled that the civil standard of proof is relevant to criminal law because most preliminary factual issues are determined according to that standard.

The SCC does not refer to the NZSC decision, and the NZSC majority in Z barely refer to the Canadian law (but to the extent that they do, at 117, they mention the flexible approach in professional disciplinary proceedings that has now been overruled).

In FH v McDougall the Court declares (at 40) the following propositions to be wrong:

"(1) The criminal standard of proof applies in civil cases depending upon the seriousness of the allegation;

"(2) An intermediate standard of proof between the civil standard and the criminal standard commensurate with the occasion applies to civil cases;

"(3) No heightened standard of proof applies in civil cases, but the evidence must be scrutinized with greater care where the allegation is serious;

"(4) No heightened standard of proof applies in civil cases, but evidence must be clear, convincing and cogent; and

"(5) No heightened standard of proof applies in civil cases, but the more improbable the event, the stronger the evidence is needed to meet the balance of probabilities test."

No, the only standard of proof in civil cases is the balance of probabilities, there is no intermediate standard applicable on some occasions. The evidence must always be scrutinised with great care. To meet the standard of proof evidence must always be clear, convincing and cogent. And the probability or improbability of the event does not affect the strength of the evidence needed, as a matter of law, although the inherent probabilities will be taken into account by the fact finder.

I suppose that the conflict can be resolved by reading the majority in Z as holding that the flexibility they found in the civil standard of proof is not "flexibility as a matter of law" but rather it is flexibility in the sense that the inherent probabilities, as a matter of practical experience, will always be taken into account by the fact finder. This reading has support at para 105 of Z, but the joint judgment has chosen unfortunate terminology in its insistence on "a flexibly applied civil standard of proof". There are clear indications that the majority in Z intended to support the proposition - and this is put as a rule (para 112) - that serious consequences justify a requirement for evidence of heightened cogency, a proposition rejected in FH v McDougall.

Elias CJ addressed (in dissenting) the different approaches in Z as follows (28):

"It is often said that more grave allegations are less likely to be true and require more in the way of evidence before the trier of fact will be satisfied [footnote: Hornal v Neuberger Products Ltd [1957] 1 QB 247 at p 266 per Lord Morris; Budget Rent ACar Ltd v Auckland Regional Authority [1985] 2 NZLR 414 at p 425 (CA) per Somers J.] I have some doubts as to the extent to which experience bears out the proposition, but in any event it is clear that its application turns on human experience and the particular context, as Lord Nicholls made clear in Re H. Statements such as these have however caused confusion when applied, not to the inherent probabilities which any decision-maker necessarily weighs, but to the standard of proof [footnote: As described by Lord Hoffmann in Re B at para [12]]. The confusion has led to judicial statements which suggest that the standard of proof is itself "flexible", an unfortunate and inaccurate notion. Nor do I think matters are improved by the suggestion that it is not the standard but its application that is "flexible". "Flexibility" is a term I think best avoided in the context of proof, despite its impressive pedigree [footnote: Lord Bingham in B v Chief Constable of Avon and Somerset Constabulary [2001] 1 WLR 340 at pp 353 – 354 (HL); Lord Scarman in R v Secretary of State for the Home Department, ex pKhawaja [1984] 1 AC 74 at p 113; Lord Nicholls in Re H at p 586]. Proof is made out whenever a decision-maker is carried beyond indecision to the point of acceptance either that a fact is more probable than not (if the standard is on the balance of probabilities) or that he has no reasonable doubt about it (if the standard is proof beyond reasonable doubt) [footnote: Miller v Minister of Pensions [1947] 2 All ER 372 at pp 373 – 374 (KB) per Denning J; Rejfek vMcElroy (1965) 112 CLR 517 at para [11]]."

It is interesting to see the senior appellate courts dealing with the same issues within a brief time span. No doubt the High Court of Australia will chip in soon.

Risk of unfairness: is the criminal standard of proof applicable?

On a day when I am feeling particularly smug, I am not going to spoil my mood by risking being in contempt of court. So my comments on Solicitor General v Fairfax New Zealand Ltd 10/10/08, Randerson CHCJ, Gendall J, HC Wellington CIV-2008-485-705 will have a narrow focus.

The plaintiff alleged that the respondents were in contempt of court because they had published information that gave rise to a real risk of interference with the due administration of justice by compromising the fair trial rights of the persons who currently face criminal charges. This allegation was held not to be made out (134):

"We conclude that the Solicitor-General has not proved beyond reasonable doubt that, as a matter of practical reality, the actions of the respondents in publishing the Fairfax articles caused a real risk of interference with the administration of justice by compromising the fair trial rights of the accused. ..."

Here we have a compound of several ideas: the right to a fair trial, the standard to which it must be proved that a trial would not be fair, a real risk of interference with the right, and the standard of proof that there is that real risk.

In Chamberlains v Lai (blogged here 11 September 2006) the Supreme Court noted that a "real risk" of the loss of a more favourable verdict is a lesser standard of proof than the balance of probabilities. A person could show that such a risk existed by adducing relatively slight evidence, and the sensitivity of the court to the issue would operate to put the burden on the other side of excluding that risk.

The present case concerns the risk that a trial could be unfair. The right to a fair trial is absolute (Condon v R, blogged here 24 August 2006) and the fairness of the trial is a matter for judicial sensitivity. Yet here, arguably, the High Court has diminished that in its formulation of its conclusion.

Certainly, the prosecution must prove its allegation of contempt to the standard of beyond reasonable doubt. What that means here is that the contents of the publications and the identities of the publishers must be proved beyond reasonable doubt. The publications in question must be of the kind that could give rise to a real risk of trial unfairness, otherwise they are irrelevant to the proceedings. But once that relevance test – which does not carry a particular standard of proof - is met, the only live issues for proof beyond reasonable doubt, given that the contents of the publications are not able to be disputed, are the identities of the publishers.

In Gisborne Herald Co Ltd v Solicitor-General [1995] 3 NZLR 563; (1995) 13 CRNZ 244, Richardson P, giving the judgment of the Court of Appeal, held:

“The present rule is that, where on the conventional analysis freedom of expression and fair trial rights cannot both be fully assured, it is appropriate in our free and democratic society to temporarily curtail freedom of media expression so as to guarantee a fair trial.”

In the present case the Full Court departed from this by requiring the prosecution to show beyond reasonable doubt that the publication prejudiced trial fairness: this is not a guarantee of a fair trial, it would permit trials that are probably going to be unfair as long as they would not beyond reasonable doubt be unfair.
See also my blog for 19 July 2005,"Proof and Risks".

Reviewing judge alone verdicts

In two decisions yesterday, the second of which applied the first, the Supreme Court of Canada explained when the reasons for verdict at a judge alone trial are sufficient to survive review on appeal: R v REM [2008] SCC 51 and R v HSB [2008] SCC 52 (both 2 October 2008).

The key to assessing the adequacy of reasons is appreciation of the functions of such reasons and to apply those to the context of the instant case. These are set out in REM at paras 11 and 12, and the Court, in a unanimous judgment in each case delivered by McLachlin CJ, also referred to R v Sheppard [2002] SCC 26 at 46, 50 and 55 for a fuller statement of them.

An overview of what is required of an appellate court is set out in REM at 55-57:

"[55] The appellate court, proceeding with deference, must ask itself whether the reasons, considered with the evidentiary record, the submissions of counsel and the live issues at the trial, reveals the basis for the verdict reached. It must look at the reasons in their entire context. It must ask itself whether, viewed thus, the trial judge appears to have seized the substance of the critical issues on the trial. If the evidence is contradictory or confusing, the appellate court should ask whether the trial judge appears to have recognized and dealt with the contradictions. If there is a difficult or novel question of law, it should ask itself if the trial judge has recognized and dealt with that issue.

"[56] If the answers to these questions are affirmative, the reasons are not deficient, notwithstanding lack of detail and notwithstanding the fact that they are less than ideal. The trial judge should not be found to have erred in law for failing to describe every consideration leading to a finding of credibility, or to the conclusion of guilt or innocence. Nor should error of law be found because the trial judge has failed to reconcile every frailty in the evidence or allude to every relevant principle of law. Reasonable inferences need not be spelled out. For example if, in a case that turns on credibility, a trial judge explains that he or she has rejected the accused's evidence, but fails to state that he or she has a reasonable doubt, this does not constitute an error of law; in such a case the conviction itself raises an inference that the accused's evidence failed to raise a reasonable doubt. Finally, appellate courts must guard against simply sifting through the record and substituting their own analysis of the evidence for that of the trial judge because the reasons do not comply with their idea of ideal reasons. As was established in Harper v. The Queen, 1982 CanLII 11 (S.C.C.), [1982] 1 S.C.R. 2, at p. 14, "[a]n appellate tribunal has neither the duty nor the right to reassess evidence at trial for the purpose of determining guilt or innocence. ... Where the record, including the reasons for judgment, discloses a lack of appreciation of relevant evidence and more particularly the complete disregard of such evidence, then it falls upon the reviewing tribunal to intercede."

"[57] Appellate courts must ask themselves the critical question set out in Sheppard: do the trial judge's reasons, considered in the context of the evidentiary record, the live issues as they emerged at trial and the submissions of counsel, deprive the appellant of the right to meaningful appellate review? To conduct meaningful appellate review, the court must be able to discern the foundation of the conviction. Essential findings of credibility must have been made, and critical issues of law must have been resolved. If the appellate court concludes that the trial judge on the record as a whole did not deal with the substance of the critical issues on the case (as was the case in Sheppard and Dinardo
[[2008] SCC 24]), then, and then only, is it entitled to conclude that the deficiency of the reasons constitute error in law."

In each of the present cases the British Columbia Court of Appeal had erred, in REM by approaching the case from a defence perspective of scepticism, and in that case and in HSB by substituting its own credibility findings for those of the respective trial judges.

The Court did not refer to its own rather weak decision in R v Jackson [2007] SCC 52 (blogged here 6 December 2007) perhaps giving us reason to think that case is best forgotten, and there would have been value in considering AK v Western Australia (blogged here 27 March 2008) for a comparison with the appellate approach to miscarriage of justice.

Considerations such as these will be increasingly relevant as judge alone trials become more frequent, as they will when judges have the power to order trial without jury for cases that are long (more than 20 days) or complex (so that a jury would be unlikely to be able to perform its function effectively), or where there are reasonable grounds to apprehend intimidation of current or future jurors. Such provisions are due to come into force in New Zealand on 25 December 2008 pursuant to s 361D and E of the Crimes Act 1961.

Judges as scholars

Who are top of the heap, judges or academics?

Should academics be allowed to slap down judges for their indulgence in scholarship when they should be simply deciding cases? Do judges have a responsibility to bring realism to the dreamy theorising of academics?

The senior judges write much of their decisions for the benefit of law students and lawyers. Who else is going to read them, especially when there is no right of appeal? Legal education does not stop at the end of the final classroom session. Indeed, the learning of the law barely starts at law school, even though significant skills and knowledge may be acquired there.

Academics learn the law primarily from judges. To the limited extent that they can bring anything new to the law, academics may be able to point to ways of resolving the contradictions that can arise in case law, by devoting time to research. Research in law usually means looking at the way the same problem has been addressed or avoided by courts in other jurisdictions. If academics expect judges to pay attention to their research, can they criticise judges for doing research of their own? There is no monopoly on the application of legal skills.

Thoughts such as these are prompted by the recently published study by Professor James Allan, Professor Grant Huscroft, and PhD candidate Nessa Lynch, "The Citation of Overseas Authority in Rights Litigation in New Zealand: How Much Bark? How Much Bite?" (2007) Otago Law Review Vol 11 No 3, downloadable here. See also this article in the New York Times, which refers to that article and to several other studies.

A spectacular example of judicial scholarship, which some might say was at the expense of judgment (I don't know whether it was or was not), occurred last week in the New Zealand Court of Appeal: Lab Tests Ltd v Auckland District Health Board [2008] NZCA 385. This is not a criminal case, but don't be alarmed. Just look at the judgment of Hammond J, which starts at para 348. The approach here is cheerful, learned, and above all interesting. Hammond J, a former Dean of the Auckland University Law School, mentions - without condescension - the conflicting views of academic commentators. This is apparently effortless scholarship, and will surely be the envy of academics. It demonstrates that legal learning can be great fun.

A cynical view of judicial scholarship seems to underlie the Otago study cited above :

"Increasingly judges act as academics, giving speeches, participating in conferences, writing law review articles, and contributing to books – if not writing them... Like academics, they are keen to expand their influence, and that of their courts. Rights are a language that is spoken internationally, and judges are writing not simply for domestic but also for overseas audiences – especially other judges and academics, all of whom may promote and further disseminate their decisions. It is not only individual judges who are keen to promote their decisions internationally. Courts are keen to promote and disseminate their decisions internationally as well...."[footnotes omitted] (p 11).

This seems to be a human failing (p 12):

"In all likelihood, personal relationships also play an important role; friendships between judges from different countries – perhaps formed through meetings at academic and judicial conferences – may lead them to cite the decisions of their respective courts. Growth in the use of court clerks is another contributing factor. Left on their own, many judges would have little time or inclination for the sort of research that rights-based internationalism demands. Increasingly, however, courts utilize clerks and executive-type legal officers, many of whom are recent law school graduates and, as a result, likely to be conversant in overseas case law, if not enthusiastic proponents of it."

On this appraisal, the motives for judicial scholarliness are influence-seeking, and friendship-forging, assisted by subservience to law clerks. Winning friends and influencing people might be by-products, but I doubt that anything other than the joy of scholarship is the real motivation for doing what academics would like to think academics do best.

And why shouldn’t judges be scholars? They have plenty of time for it, as do lawyers. At least Dr Johnson thought they did, as noted by Boswell (Life of Johnson, narrative for Thursday 6th April 1775):

“No, Sir, there is no profession to which a man gives a very great proportion of his time. It is wonderful, when a calculation is made, how little the mind is actually employed in the discharge of any profession. No man would be a judge, upon the condition of being totally a judge. The best employed lawyer has his mind at work but for a small proportion of his time; a great deal of his occupation is merely mechanical.”

How much have you got?

The presumption of innocence refers to two things: the burden of proof and the standard of proof. It means that the burden of proof is on the prosecution, and the standard of proof is beyond reasonable doubt. The presumption of innocence applies at a criminal trial when the ultimate issue is whether the defendant is guilty.

The presumption of innocence is not the same thing as the right to a fair trial, although in some generalised conceptions of fair trial the presumption of innocence is regarded as a requirement for trial fairness. But it need not be, and in any event the presumption of innocence has exceptions.

The most obvious exception to the presumption of innocence is that if the defendant is relying on the defence of insanity, he must prove that on the balance of probabilities he was insane at the relevant time. This exception is justified on the basis that he will know more about his state of mind than the prosecution, so he should have the burden, and, as it seems unfair to require him to prove insanity to the standard of beyond reasonable doubt, the lesser standard of the balance of probabilities applies.

The law has not regarded the need to raise a reasonable doubt on the issue as a standard of proof. There are only two standards of proof: beyond reasonable doubt, and the balance of probabilities. See, for example Z v Dental Complaints Assessment Committee (blogged here 25 July 2008) and the Index to this site, and the entries under the heading "Standard of proof", for exceptions to this generalisation.

Another illustration of an exception to the presumption of innocence is where, on proof that he had a particular quantity of illicit drug in his possession, the defendant must prove on the balance of probabilities that he did not have an intention to supply it to another person, if he is to avoid conviction for possession of the drug for supply. This exception is more controversial than the insanity one, because the prosecution should have access to evidence of the defendant's dealings, and should be able to prove circumstances indicating an intention to supply the drug. That is not particularly different from requiring the prosecution to prove the mental elements of possession of the drug. So this exception to the presumption of innocence may be thought to lack sufficient justification in policy terms and may be a breach of the right to be presumed innocent. See Hansen v R [2007] NZSC 7 (blogged 20 February 2007). In the UK the raising of a reasonable doubt has been held to be "proving" absence of a mental element: R v DPP ex p. Kebilene [2002] AC 326, so that amounts to recognition of a third standard of proof.

Sometimes a reverse onus of proof is applied to the issue of the magnitude of a confiscation order in proceeds of crime applications. This situation existed in Grayson and Barnham v United Kingdom [2008] ECHR 877 (23 September 2008). The focus here was on whether a provision requiring the applicants to show on the balance of probabilities that they did not have sufficient assets to meet a confiscation order, the amount of which corresponded to the court's assessment of their benefits from their respective offending, was in breach of the fair trial requirement in Art 6§1 of the Convention. This was important, not only because of the amount of money involved, but also because in default of payment prison sentences which were quite hefty were imposed.

The Strasbourg court has long regarded the presumption of innocence as part of the general notion of a fair hearing: Saunders v. United Kingdom, judgment of 17 December 1996, Reports of Judgments and Decisions 1996-VI, § 68. Also, proceedings of this nature are treated as analogous to sentencing: Phillips v. United Kingdom (no. 41087/98, §§ 35 and 39, ECHR 2001-VII.

Well, in the present case it was the applicants (the original accuseds) who had the means of proving their present inability to pay. The Court quoted Lord Steyn, with whom the other Law Lords agreed, in R v Rezvi [2002] UKHL 1, who approved the following reasons given by the Court of Appeal supporting the exception to the presumption of innocence in this context:

"(a) It is only after the necessary convictions that any question of confiscation arises. This is of significance, because the trial which results in the conviction or convictions will be one where the usual burden and standard of proof rests upon the prosecution. In addition, a defendant who is convicted of the necessary offence or offences can be taken to be aware that if he committed the offences of which he has been convicted, he would not only be liable to imprisonment or another sentence, but he would also be liable to confiscation proceedings.

(b) The prosecution has the responsibility for initiating the confiscation proceedings unless the court regards them as inappropriate ...

(c) There is also the responsibility placed upon the court not to make a confiscation order when there is a serious risk of injustice. As already indicated, this will involve the court, before it makes a confiscation order, standing back and deciding whether there is a risk of injustice. If the court decides there is, then the confiscation order will not be made.

(d) There is the role of this court on appeal to ensure there is no unfairness."

Those backstop matters – the appropriateness of the proceedings, the avoidance of a serious risk of injustice, and the appellate court's role of ensuring there is no unfairness – were reflected in the ECtHR's conclusion (49):

"The Court agrees with the judgments of the Court of Appeal in the instant cases ... that it was not incompatible with the notion of a fair hearing in criminal proceedings to place the onus on each applicant to give a credible account of his current financial situation. In each case, having been proved to have been involved in extensive and lucrative drug dealing over a period of years, it was not unreasonable to expect the applicants to explain what had happened to all the money shown by the prosecution to have been in their possession, any more than it was unreasonable at the first stage of the procedure to expect them to show the legitimacy of the source of such money or assets. Such matters fell within the applicants' particular knowledge and the burden on each of them would not have been difficult to meet if their accounts of their financial affairs had been true."

There was therefore no violation of the right to a fair trial arising from the reverse onus. The applicants had argued that they were in the unfair position of having to prove a negative, and (for one of the applicants) that the assumptions made by the lower court when it assessed the benefit obtained from offending continued to operate as assumptions when the court decided the ability to pay. The point seems to be that not only did the applicant have to prove inability on the balance of probabilities, but also he had to do so in the face of an additional assumption against him. This was not explored by the Court.

The argument raises two concerns: did the additional assumption against the defendant, assuming it existed, mean that unfairness occurred in the sense that the standard of proof he had to satisy was raised by requiring him to adduce more cogent evidence than he otherwise would have had to adduce; or, was there unfairness arising from a bias by the court against the defendant, which again required him to adduce more cogent evidence that should have been the case? European law asserts that the standard, the balance of probabilities, does not change with the seriousness of the consequences of the decision (see blog discussing Z, above). However, here there was nothing to indicate that the courts in the case of each of the applicants had applied the wrong standard or that they were biased.

More on standards of proof of preliminary facts

In Canada it is necessary that compliance with procedures for interviewing young people be established beyond reasonable doubt: R v LTH [2008] SCC 49 (11 September 2008). The various protections enacted in s 146(2) of the Youth Criminal Justice Act, SC 2002, c. 1 (which does not specify a standard of proof for compliance) are intended to ensure that statements made by young people are reliable (38), and proof of waiver of them is also required to be established beyond reasonable doubt (39).

The balance of probabilities is the standard applicable in New Zealand in respect of preliminary facts such as compliance with the right of a person charged with an offence to be brought as soon as possible before a court: R v Te Kira [1993] 3 NZLR 257, and compliance with the right of persons arrested or detained to consult and instruct a lawyer without delay: Police v Kohler [1993] 3 NZLR 129, and the same applies to young people: R v S (CA220/97) (1997) 15 CRNZ 214. There is some fudging of the standard, as it is called the balance of probabilities having regard to the gravity of the particular issue:

“The first issue is whether the Crown must prove the facts necessary to establish admissibility of a statement under s 221(2) [Children, Young Persons, and Their Families Act 1989] to the criminal standard of beyond reasonable doubt. In the High Court, Paterson J appears to have adopted this higher test, citing the judgment of Thorp J in R v Fitzgerald unreported, 30 October 1990, HC Auckland T183/90.

“Fitzgerald concerned compliance with the advice requirements of s 215(1)(f), which must be met before any questioning of a young person is undertaken. It was held that substantial compliance had to be proved to the criminal standard before a statement would be admitted. No argument appears to have been addressed to the issue of standard of proof. Since Fitzgerald this Court has held that in the analogous cases concerning alleged breaches of the New Zealand Bill of Rights Act 1990, the lower standard of proof, or balance of probabilities having regard to the gravity of the particular issue, is to be applied R v Te Kira ... Police v Kohler.... We see no good reason to elevate the test under this legislation [Children, Young Persons, and Their Families Act 1989,ss 221(2)(b), 222(4)], and to add a second exception to the ordinary rule governing incidental trial issues. Proof beyond reasonable doubt is therefore not required.”

We have noted recently the discussion by the Supreme Court of what relevance the gravity of the consequences has to the standard of proof when the standard is the balance of probabilities: Z v Dental Complaints Assessment Committee, blogged 25 July 2008.

There are good reasons to complain that the fudged balance of probabilities should be replaced by the standard of proof beyond reasonable doubt. Particularly so, because in Kohler Cooke P (as he then was), delivering the decision of the Court of Appeal, said:

“The question of the standard of proof required of the prosecution to establish a waiver was raised in argument. Some American and Canadian judgments speak of a "heavy burden'', a "high standard of proof'', a "very high standard''. See for example Miranda at p 475; Moran v Burbine 475 US 412, 439, n 9, 450-452 (1986); and R v Manninen (1987) 34 CCC (3d) 385, 393. In Te Kira this Court determined that, for questions of fact concerning alleged Bill of Rights breaches, once the situation is such that the onus falls on the prosecution to negative a breach, the appropriate standard is the balance of probabilities but with the gravity of the issue borne in mind. We doubt whether in practice this will lead to results different from those that would be reached in Canada and the US. The Te Kira standard should apply on the issue of waiver as well as on other factual issues under the Bill of Rights Act. As already mentioned, the prosecution must fail by that standard here.”

If the results are not going to be different, what is the problem with requiring proof beyond reasonable doubt?

Well, in NZ the position is determined by the Evidence Act 2006, which was drafted in the light – if that's the right word – of the authorities mentioned above. The Act specifies – with the unduly fussy particularity that is currently regarded as good draftsmanship – that a finding that evidence was obtained improperly is to be made on the balance of probabilities: s 30(2)(a).

I can't help wondering whether Lord Cooke would have agreed with the decision of the House of Lords in In re B (blogged here 12 June 2008). I think he would have, and in doing so he would have revised the approach he took in Kohler, above. My current take on this is that the Evidence Act 2006 impedes progress on this point, as has the majority decision in Z, above.

This is not to say that all preliminary facts (those which must be established before other evidence becomes admissible) should be proved to the standard beyond reasonable doubt. There is nothing illogical in requiring the balance of probabilities for these, even where the ultimate issue (guilt) must be proved beyond reasonable doubt. The criteria for admissibility are separate from the criteria for proof of guilt. The choice of standard for proof of preliminary facts is a matter of policy, and in the present context it reflects the importance that the court attaches to compliance with the rights of the accused. In Kohler, above, the Court did not discuss the competing policies or give extended reasons for its choice. Similarly, in R v S, above, the Court simply saw “no good reason” for applying the standard of beyond reasonable doubt, again without policy analysis. Even the Privy Council has decided the issue without extended policy analysis: Smith v R (Jamaica) [2008] UKPC (blogged here 27 June 2008), applying the criminal standard. The reasons for the choice are a mystery.

I say “even” the Privy Council, because the Law Lords, who write their opinions without the assistance of law clerks, are least likely to use a style redolent of the honours dissertation. Posner, in “How Judges Think” (2008), makes this interesting observation (p 221):

“Students are taught to approach judicial opinions as if every word were written by the nominal author – that is, the judge – and the effect is to imbue them with a legalistic outlook, an effect reinforced by their youth (to which ... algorithmic thinking as distinguished from pragmatic or legislative thinking is more congenial than it is to older persons) and by an understandable desire to believe that their steep law school tuition is buying them a set of powerful analytic tools. When they become law clerks it is natural for them to write opinions designed to provide legalistic justifications for their judges' votes. They thus contribute to the mystification of the next student generation.”

In the absence of legalistic justification for the choice of standard of proof of preliminary facts, other than regarding earlier cases as precedents that should (but why?) be followed, there is a need for policy justification, but legal education does not develop policy reasoning.

I'm bad, I'm bad!

“I am so bad that you can't rely on my confession”: like the Cretan liar, the accused in R v Bonisteel [2008] BCCA 344 (9 September 2008) presents a paradox.

He had boasted about his criminal past in order to gain acceptance into a criminal organisation. At least, that's what he thought the organisation was. In fact he was talking to undercover police officers. They were trying to get him to confess to the murder of two girls who had died from stab wounds. In pretending to be potentially violent and a person not to be messed with, the main interviewing undercover officer questioned the accused in ways that the accused claimed at trial were oppressive. As a result the accused said he did the killings, but he was inaccurate in his account of the wounds inflicted and in his description of the girls.

If his confession to the killings was admissible at trial, his defence tactic was to say that he was lying in order to gain acceptance into the organisation. The bad things he said about himself, including accounts of other offending and prison sentences served, were said with that objective. The reasons that the gang should think he was bad were the same as the reasons the jury should think he was not bad.

There are some unsatisfactory aspects to the British Columbia Court of Appeal judgment (a judgment by Levine J, with which Lowry and Bauman JJ agreed): see CriminalReview.ca. In particular, these concern whether the defence should have been allowed to call expert evidence on false confessions (67-71). There is a floodgates danger, of course, in allowing such evidence.

What is odd is the Court's treatment of the admissibility of bad character evidence. This came first in order of points considered, whereas one would have expected it to come after the oppression point. If the undercover officer's evidence was not obtained through oppression, or other impropriety, then should it be edited to exclude reference to the accused's bad conduct on other occasions?

Instead, the Court approved the trial judge's reasoning (48):

“Thus, the trial judge did not expressly consider editing the statements because his analysis led him to the conclusion that the prejudicial portions were relevant to an issue in the case – the truthfulness of the appellant’s confession – their probative value outweighed their prejudicial effect, and the prejudice could be dealt with by a prophylactic warning. In other words, the prejudicial evidence was neither 'irrelevant' nor 'unnecessarily prejudicial to the accused', so the duty to edit did not arise.”

This led the Court to address the decision on whether a warning to the jury would be an adequate substitute for exclusion of the evidence, and it held that the judge had correctly decided that a warning would suffice.

There is no suggestion in the Court's judgment that the prosecution was able to call evidence of the accused's character other than what the accused had said to the undercover officer. The focus was on the relevance of that evidence to the truth of the confession, that is, to its reliability. However reliability does not cure oppression or impropriety which are logically the prior considerations.

Well, if the Court had addressed the confession admissibility issue first, would it have decided that it had been correctly ruled admissible?

Probably yes: the accused had put himself in a position where he must have realised that the sort of conduct that persuaded him to confess (or more likely, to boast) was likely to occur. In the absence of oppression, the next question is whether there was improper police conduct such as to justify exclusion of the confession on public policy grounds. Unlikely, as far as one can tell from the judgment (86-94). So, the confession was admissible. What then of the bad conduct evidence?

The answer is that, once the confession became admissible, the defence would want the bad conduct evidence in to support its claim that the confession was just big talk. Boasting about his criminal past was part of the accused's tactic to gain acceptance into what he thought was a criminal organisation. His defence was, in effect, “I'm bad, but not that bad.”

Business as usual

Like everyone else, I spend much of my time looking out the window. Except for blind people of course. And those in cells without views. My view is of a forest that looks much as it must have millions of years ago, minus the large and dangerous fauna. I say millions of years with some confidence, as this part of the North Island apparently dates from the late Jurassic. The exasperations of the law have caused me to look at this scene many times. One of the things I wonder about, and have done so today, is why so many criminal appeals are concerned with points that should be obvious.

Vast numbers of criminal appeals are futile, or result from errors that should never have happened. If there is something wrong with this, who is to blame? Are lower courts wilfully contrary or hopelessly dull witted? Are counsel afraid to advise against appeals? Richard Posner in “How Judges Think” (2008, Harvard University Press) says that the high number of criminal appeals is due to appellants not having to pay any costs.

Should lawyers take more responsibility for making decisions – or, more properly, giving firm advice – based on the prospects of success on appeal? Do insurance policies protecting counsel from liability for negligence require that every avenue be explored? Should there be some penalty on judges who make fundamental errors (send them back for a stint in the jurisdiction of first instance)?

It is very rare in a busy criminal practice to come across a point that is genuinely worth appealing. At least it is if the case has been properly handled in the lower court. You would not think so if you got your impressions from reading appellate cases. Those may suggest that every case bristles with genuine appeal points. But they are a very small proportion of the number of cases that are commenced.

But enough of the preamble. Today the High Court of Australia decided that supply of cocaine is likely to seriously endanger the health of the end user: Gedeon v Commissioner NSW Crime Commission [2008] HCA 43. I look out the window and sigh.