Suppressing suppression orders: Farish v R [2024] NZSC 65

I am going to comment on this case without mentioning Kafka. Franz Kafka. 1883-1924. Author of The Trial, concerning prosecution for an undisclosed crime.

Anyone who is less respectful of the law than I am might thing of Kafka when they read the order issued by the Court in the publicly available judgment in Farish v R [2024] NZSC 65:

“[82] We make a permanent order … prohibiting publication of the following:

(a) any evidence and submissions in this proceeding relating to the nature of the risk [REDACTED] and the information relevant to that risk [REDACTED]; and 

(b) [REDACTED].”

Does such an order meet the requirement of the rule of law that the law should be ascertainable? Strictly speaking, a ruling is not a "law", not a legal norm effective as a precedent, but because it applies to future behaviour it needs to be ascertainable by more people than those directly involved in the case. [1]

The order, to be understood, needs to be read in its context, but in this case the context is in the Court’s judgment that is not published.

In a more helpful part of its judgment, the Court discusses s 205 of the Criminal Procedure Act 2011 which provides for suppression of evidence and submissions (see [25]-[33]).

Particularly noteworthy is the way the Court addressed the problem of whether suppression here might cause the public to think that the judicial system was favouring its own (the applicant for suppression here being a judge, who was not a defendant and who was not involved in the offending).

The public should be the judges of what happens in the courts [73], referring to Lady Hale P in Dring (on behalf of the Asbestos Victims Support Groups Forum UK) v Cape Intermediate Holdings Ltd (Media Lawyers Association intervening) [2019] UKSC 38, [2020] AC 629 at [42]-[43]. The Court concluded that “the interests of justice do not require that the appellant accept a serious risk to personal safety so that a court can avoid giving the false impression that she is being afforded privileged treatment.”

__________________________

[1] See John Gardner, Law as a Leap of Faith (2012, OUP) p 193.

Fundamental but simple points in two important cases: Thornell v Jones USSC 22-982 and A (SC70/2022) v Minister of Internal Affairs [2024] NZSC 63

Two recent cases highlight a couple of unrelated but fundamental points.

When has a defendant not had effective legal representation?

In Thornell v Jones USSC 22-982, 30 May 2024 the Court summarised the law on appeals against sentence on the grounds of ineffective assistance of counsel, which had been stated in Strickland:

“To succeed on such a claim, a defendant must show that counsel provided a “deficient” performance that “prejudiced” him. Strickland, 466 U. S., at 687. When an ineffective-assistance-of-counsel claim is based on counsel’s performance at the sentencing phase of a capital case, a defendant is prejudiced only if “there is a reasonable probability that, absent [counsel’s] errors, the sentencer . . . would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Id., at 695. “A reasonable probability is a probability sufficient to undermine confidence in the outcome. That requires a substantial, not just conceivable, likelihood of a different result.” Pinholster, 563 U. S., at 189 (citation and internal quotation marks omitted). This standard does not require a defendant to show that it is more likely than not that adequate representation would have led to a better result, but “[t]he difference” should matter “only in the rarest case.” Strickland, 466 U. S., at 697. To determine whether a prisoner satisfies this standard, a court must “consider the totality of the evidence before the judge or jury”—both mitigating and aggravating Id., at 695.”

Suspicion and belief are different states of mind

And in A (SC70/2022) v Minister of Internal Affairs [2024] NZSC 63 at [41] the fundamental distinction between belief and suspicion was accepted and applied:

"Clause 2(2) requires that the Minister believes on reasonable grounds, not suspects. Both counsel for the appellant and the special advocate emphasised this. In R v Williams, a case dealing with applications for search warrants, the Court of Appeal explained the difference between these standards in these terms: [Footnote 23: R v Williams [2007] NZCA 52, [2007] 3 NZLR 207 per William Young P and Glazebrook J (citations omitted). Hammond J agreed at [258].]

[213] Having “reasonable grounds to believe” … is a higher standard to meet than “reasonable ground to suspect” … Belief means that there has to be an objective and credible basis for thinking that a search will turn up the item(s) named in the warrant, while suspicion means thinking that it is likely that a situation exists. The issuing officer must hold the view that the state of affairs the applicant officer is suggesting actually exists.”

This case was a successful appeal against the dismissal of an application to review the Minister’s cancellation of the applicant’s passport. The Supreme Court declared that the Minister’s decision was unlawful and invalid [159] and awarded her costs in relation to the appeal [161]. There is a good deal more of interest in this case concerning the interpretation of the relevant anti-terrorist legislation. The special advocate procedure was used and a significant proportion of the proceedings occurred in closed court [5]. The Supreme Court issued two judgments, this public one and another one only accessible in narrowly defined circumstances [7].

The (ir)relevance of a sexual relationship: R v T.W.W., 2024 SCC 19

One of the interesting things about appellate court benches that consist of a large number of judges is that a few dissenters can produce a convincing judgment.

A split of 7-2 in the Supreme Court of Canada in R v T.W.W., 2024 SCC 19 illustrates this.

There are procedural points addressed in this case that are of great interest to Canadians, but that I will not discuss here.

Rather, it is the difference over whether the defendant’s claim, that a sexual encounter was not rape because the complainant consented, could be supported by evidence that he and she had had consensual sexual intercourse the previous day.

Everyone agrees that it is wrong to reason that prior consent of itself supports present consent. And it is wrong to reason that prior consent of itself shows that the complainant’s denial of present consent is a lie.

But what if a complainant says that there was no relevant relationship with the defendant before the alleged rape? For example, as in this case as assessed by the dissenting justices Côté and Moreau JJ, although they had been married for 20 years and they still lived in the same house, but now separately, the complainant claimed that their relationship had broken down to the point where she no longer consented to sexual conduct with the defendant. The complainant had implied that she would not have consented to sex with him during their separation [140].

To show that this denial of sexual conduct was incorrect, the defendant should - perhaps you agree - be able to adduce evidence of their having had consensual sexual intercourse the day before the alleged rape [122], [127]. Indeed, the complainant had told the police of that consensual occasion, so it was unlikely to have been a disputed occurrence if it had been disclosed in evidence at trial [127].

The evidence would not be used in this situation to support one of the “twin myths”, namely that here the occurence of sexual intercourse the previous day made it more likely that she would have consented on the present occasion, or that it made her less likely to be telling the truth. Instead, it contradicted her claim of what one might call sexual separation.

Why couldn’t the Supreme Court correct what might appear to be the serious error of not admitting evidence of the previous day’s consensual sexual intercourse?

The majority said that the pre-trial application to adduce the evidence had lacked sufficient specificity [27]-[28], [36], [38]. And that at trial the defence was that the alleged act of intercourse, the rape, had not occurred at all [8]. If it hadn’t occurred, no question of its being consensual could arise. And further, the complainant had not been inconsistent because her denial of a continuing romantic relationship was not a denial of a sexual relationship [40], so evidence of the consensual act of intercourse during the separation would have had little probative value [42]. Appeals to the Supreme Court against conviction have limited scope [46], and here the question whether the trial judge should have re-visited the pre-trial ruling is not an issue for consideration on this appeal [48], but in any event the issue would not have called for reconsideration at trial in this case [53].

Overall, the majority seem to be framing the relevant issue according to what they consider the defence position was at trial, while the minority frame the issue according to what they perceive the complainant to have been saying.

I have the impression that this was one of those trials where, because of the background of a long-standing romantic relationship, both the complainant and the defendant were having difficulty in being clear about what they were saying.

The persistence of dangerousness: Brown v United States 22-6389, May 23 2024

A potential sentence for an offence may be increased if the offender has a record of serious offending.

The policy would be to reflect the social danger posed by such an offender, evidenced by persistent serious criminality.

What if the offences for which an offender has convictions are amended by subsequent legislation to make them less serious? They might no longer make the offender eligible for the increased sentence for the latest offence.

In some jurisdictions, legislation about the interpretation of legislation [1] may   make the answer fairly clear.

In Brown v United States 22-6389 (May 23, 2024) the Supreme Court of the United States divided over, if I might put it very broadly, the effect of the change in seriousness of some previous drug convictions for the purposes of sentencing for a firearms offence.

The majority (Alito J, joined by Roberts CJ, Thomas, Sotomayor, Kavanaugh and Barrett JJ) held that the policy of the legislation under which the present sentences were imposed was to reflect the increased social danger posed by persistent offenders, and that this danger was present regardless of the legislature’s reduction in the seriousness of the previous offending.

By way of example, the majority referred to the fact that after Prohibition, the legalization of alcohol “did not by any means ensure that these bootleggers would take up legitimate jobs … many of them simply shifted to other illegal enterprises.”

The position of the minority (Jackson J, joined by Kagan J and in part by Gorsuch J) was that the law is applicable as at the time of the present offending. This, they said, was clear from the plain words of the legislation under which the present sentences were imposed.

The minority added, in a part not joined by Gorsuch J who otherwise joined in the dissent, that the majority had not explained how the future dangerousness of an offender is best assessed by outdated assessments of the seriousness of previous offending. Furthermore, the majority’s interpretation did not include present offenders whose earlier offending would now be assessed as serious but which was not serious when it occurred.

And in a footnote Jackson J described the reference to bootleggers as “strained”, noting that many had successful legitimate careers, and in any event there was no evidence that Congress drew the same lessons from the Prohibition era when considering the present legislation.

The different approaches in this case leave us wondering whether dangerousness is best assessed by act or by attitude.

_____________________________

[1] See, for example, the Legislation Act 2019 [NZ], s 35 . As its heading indicates, this gives continuing effect to powers exercised under legislation that has been repealed or amended. An example, I think, would be a conviction entered for an offence that has been amended or repealed; the conviction continues in existence, subject to contrary legislative provision. Sections 11 and 12 appear to be consistent with this. Contrary legislation might deem the earlier offences never to have happened, or might establish a procedure to “expunge” the convictions, for example as does the Criminal Records (Expungement of Convictions for Historical Homosexual Offences) Act 2018 [NZ]; for the effect of expungement, see s 9.

Morality and causal attribution: R v Lozada, 2024 SCC 18

Whether a potential causal link between the defendant’s act and the harm proscribed by an offence was displaced by an act of another person can be a difficult factual issue. It can also be thought of as a moral issue, and it was so described by the trial judge whose instruction to the jury was the subject of appeal in R v Lozada, 2024 SCC 18, [12], [40].

As this case illustrates, there are various ways in which the attribution of this causation may be addressed.

The act of the “other person” may be called an intervening act, although to some people this might seem question-begging or conculsory. It seems that here, intervening means subsequent to the defendant’s act and before the occurrence of the proscribed harm. With that sorted out, it is possible to describe the subsequent act as intervening without presupposing that the necessary causal link has been broken.

Here there was an ugly fight between two groups, during which the victim was fatally stabbed by the “other person” who was convicted of murder. The two appellants had been charged with and convicted of manslaughter as principal parties, the Crown conceding that there was no evidence that they knew or expected that anyone in their group would use a weapon.

The fight can be called “ugly”, based on the majority’s description at [23]. In that context, was there (or, should there be held to have been) a causal link between the defendants’ assaults and the victim’s death?

The fundamental legal issue is whether, for each defendant, their conduct was “a significant contributing cause” of the victim’s death. [20] In Canadian jurisprudence, one of the subsidiary questions here is whether the so-called intervening act was reasonably (objectively) foreseeable. [22] It need not have been precisely foreseeable for there still to be a causal connection between the defendants’ acts and the death, as the inquiry is about the general nature of the intervening act rather than its specific kind. [24] The focus is on the contribution of each defendant to the victim’s death, and whether it was a significant cause of that death.

Causation is case-specific and fact-driven [23]. In group assaults, in the absence of an intervening act, the actions of all assailants can contribute significantly to all injuries sustained [28]. There is no single test or measure for determining whether a particular act has broken the chain of causation [29].

It would have been wrong for a juror to have seized on one of the considerations relevant to causation and ignored the others [22]. In determining whether a specific intervening act (here, the stabbing) was in its general nature reasonably foreseeable, the jury could ask itself whether the stabbing “naturally flowed” from each defendant’s conduct, whether the stabbing was “extraordinary and highly unusual”, whether it was “directly related” to each defendant’s unlawful act, and whether the stabbing was so overwhelming as to make the acts of each defendant “merely part of the background or setting” of the death, [20] whether the conduct of each defendant rendered the victim more vulnerable to the stabbing, and whether a defendant’s acts prevented assistance being given in defence of the victim. [21]

The dissenting justices, Rowe and Jamal JJ, considered that the trial judge’s instructions to the jury had left open the incorrect conclusion that because the continuation of the assaults on the victim and the resulting risk of non-trivial bodily harm to him was reasonably foreseeable the stabbing did not break the chain of causation between the defendants’ acts and the death. [37] The rendering of the victim more vulnerable did not address the question of the foreseeability of an assault which had the general nature of a stabbing [39], and the judge’s answer to the jury’s request for a definition of a break in the chain of causation did not correct the error [40]. The judge needed to address the foreseeability of an assault which had the general nature of a stabbing, and to leave to the jury whether its general nature was a continuation of the defendants’ assaults on the victim. [42]

Special significance may be attached to a judge’s answer to a question from the jury, and appellate judges may differ on whether, when read as a whole, the jury had received a correct direction on the law. [45] 

Here, the Supreme Court of Canada split 3-2 (the majority being Karakatsanis, Martin and Moreau JJ), which makes this appeal one of those which could leave us with the uneasy feeling that, after all the appeals are over, a case may not have been conducted according to law. How can we really know?

And what about the reference to morality? This was unexplored in the appeal judgments. The judge's instruction seems to leave open the possibility that jurors might have said to themselves, "Well, we can't agree on what the judge meant about the law, except we do remember that morality is relevant. We agree that the defendants should be found guilty, so that's our verdict."

My thought is that moral considerations played their part in the development of the law, but now their role is spent. We have trials according to law, not according to people's individual moralities. However, it is impossible to prevent juries from deciding that, whatever the law may be, in the particular circumstances a conviction would be wrong. That is an important safeguard for individuals.

Unfortunately these matters of morality, which are far more interesting than the relatively straightforward law of causation, seem to have been overlooked in this case.

Rebutting an unforeseeable defence: Obian v The King [2024] HCA 18

Sometimes the prosecution can rebut evidence given by a defendant by obtaining the judge’s permission to call further evidence. The prosecutor may not have anticipated the defendant’s evidence.

Both the common law and statute may be relevant. An issue could arise as to whether the common law on the topic has been replaced by the statute to the extent of its terms, and also as to what is the extent of the statutory terms.

In Obian v The King [2024] HCA 18 the central statutory provisions were s 233(1) and (3) of the Criminal Procedure Act 2009 (Vic), set out at [10] and [11] of the Court’s unanimous judgment. Subsection (3) preserves the common law, and subsection (2) gives a separate power to permit further evidence in relation to specified Victorian criminal procedures [12]. And these specified procedures, two in number and linked by the conjunction “and”, are not exhaustive [13] so they do not replace the common law power [15]. As with the common law power, the statutory power is exercisable having regard to all the circumstances as they exist at the time the prosecutor makes the application to adduce further evidence [16].

The essential issue is whether the defendant has given evidence which the prosecutor could not reasonably have foreseen [15].

Broadly, in this case the defendant claimed to have not been involved in drug dealing, and that all he did was deliver a vehicle to a person who had not given evidence and who was not suspected of being involved in the dealing. The prosecution applied to adduce video surveillance proof that the defendant had not delivered the vehicle to that person. Should the prosecutor have anticipated that the defendant would give the explanation that he gave for the first time in court?

This was a vigorously contested point, the transcript of the argument was 70 pages [24], and although there were some innocent misstatements by counsel of the facts these did not turn out to be relevant [34].

The defendant’s explanation was not reasonably foreseeable as there were several possible exculpatory explanations that could have been given [35]-[39]. The Court agreed with the assessment by Macaulay JA in the Court of Appeal, that the defendant had, in pre-trial communications, “left an ambiguous and Delphic breadcrumb trail as to what his real defence was”. The trial judge had correctly ruled that the prosecutor could call the rebuttal evidence.

Choice of language in court: R v Tayo Tompouba, 2024 SCC 16

One of the interesting things about R v Tayo Tompouba, 2024 SCC 16 is the discussion of the criteria for a successful appeal against conviction in Canada. Another is the difference in judicial opinion over whether this case involved an error of law or a miscarriage of justice. Either way, the consequences differ. A third interesting thing is the way Canadian trials accommodate the various languages, particularly French and English.

The powers of a Canadian appellate court hearing an appeal against conviction are set out in s 686 of the Criminal Code, quoted at [53] of the majority judgment. Of note is the separation of wrong decisions on a question of law from miscarriage of justice on any ground: s 686(1)(a)(ii) and (iii).

If an appeal is advanced on the ground that there had been an error of law, and if the appellate court finds that there had been such an error, then the appellant will succeed unless the respondent can persuade the court on the balance of probabilities that the error was not substantial or that no miscarriage of justice had occurred: s 686(1)(b)(iii).

If, on the other hand, an appeal is advanced on the ground that there had been a miscarriage of justice, the appellant has the burden of convincing the appeal court that there was such a miscarriage and also that it was a miscarriage that prejudiced the appellant: s 686(b)(iv). This sort of prejudice would be the error that had an adverse effect on the outcome of the proceedings.

It is interesting (for the very few people who take an interest in such things) to compare this arrangement with the way conviction appeals are analysed in other jurisdictions. For example, here [1] our governing provision is s 232 of the Criminal Procedure Act 2011. Relevantly, we put errors of law implicitly under the rubric of miscarriages of justice, defined in s 232(4). And errors are not miscarriages of justice unless they have a consequence specified in s 232(4)(a) or (b). It is, as a matter of interpretation of the legislation, for the appellant to satisfy the court that any such consequence occurred.

So in Canada the appellant in this appeal was in a slightly better position if the error he complained of was an error of law. Here the judges differed, with Karakatsanis and Martin JJ jointly dissenting by regarding the appeal as being about a miscarriage of justice and not about an error of law.

The error, however it would be classified, was a failure at an early stage of the proceedings to draw Mr Tayo Tompouba’s attention to his right to choose which language, English or French, to have his trial conducted in. This, the third of the interesting things I mentioned at the outset, is remarkable for the extent to which Canadian trial law accommodates the two official languages [36], [38]-[42].

In New Zealand, by contrast, while Te Reo Māori is recognised in law as an official language by the Māori Language Act 2016, which is also named Te Ture mō Te Reo Māori 2016, in legal proceedings it only gives the right to speak Māori: s 7. English language does not have legal status but it is the language spoken by most people in New Zealand and may be described as the de facto official language.

Getting down to the nitty gritty, as we say in English, how did the difference in whether the error in this appeal was an error of law arise? The majority, in a judgment delivered by Wagner CJ, held that because there was a statutory duty to give an accused person information about the right to choose a language, the error was one of law [84]. It was a failure to apply a legal rule.

The minority took the view that the case law limited errors of law to errors by the trial judge, not errors that occurred, as here, at an earlier stage of the proceedings [134]. Errors of law undermine the correctness of the court’s judgment. This meant that in the minority’s opinion the error was a miscarriage of justice rather than an error of law, and so the appellant had to show that it had some effect on the exercise of his right to choose a language and he needed to produce evidence, such as an affidavit, that he was unaware of the right [135]-[136].

This suggests an interesting [2] subject for discussion: would it be preferable in legislation to avoid distinguishing errors of law from miscarriages of justice, and to require an appellant to always have the burden of satisfying the appellate court that the irregularity resulted in an adverse outcome at trial?

_____________________________

[1] By “here” I mean in New Zealand, not in the country where I happen to be writing this.

[2] I insist that it is interesting.