Try again. Fail again. Fail better.

Haven’t we seen it all before? A person wants something, they are refused with reasons. They make another request, supposedly in the light of those reasons, and they get what they want even if the second request is flawed.

This must be a psychological thing on the part of the decider. It might be a simple planning issue - whether to give permission for a helicopter landing site, for example, or a deportation surrender matter that the Minister of Justice has to decide. It’s a sort of regression: a person who is criticised on one performance of a test, is expected - even assumed -  to do better on a further similar test, but instead they tend to regress to their average performance: see Daniel Kahneman, Thinking, Fast and Slow.

A spectacular example of the latter, deportation, occurred in Kim v Minister of Justice [2019] NZCA 209. The Minister ordered deportation. The High Court on review said the Minister’s decision was flawed and referred the issue back. The Minister considered more evidence and ordered deportation again (cf, Kahneman’s repeated test), and the High Court on review said OK, you got it right this time (cf, Kahneman’s optimistic expectation of improvement), but on appeal the Court of Appeal said, no, High Court, although you were right with the first review, you got the second review wrong. (This is another regression: the High Court’s good performance of the first review was followed by a poorer performance of the second review.) The Court of Appeal ordered the Minister to reconsider the matter with particular reference to specific points (listed at [278]).

Wearing a decision-maker down with repeated applications, a practice learnt very early in life, is successful often enough for it to be an enduring behaviour. At least the Court of Appeal in Kim didn’t trouble the High Court by remitting it back, instead it left it with the Minister (a different person now) to say, “Oh, merde, not this again.”

Kim is particularly interesting for its observations on information about the criminal justice system in the People’s Republic of China, and for its recognition that a real risk that the person would be subject to an unfair trial is sufficient to refuse deportation (at [176]-[180] and [278(e)]).

Update: On 20 September 2019 the Supreme Court granted leave to each party respectively to appeal, and to cross-appeal: Minister of Justice v Kim [2019] NZSC 100.

And on 4 June 2021 the Supreme Court delivered a substantive, but not final, decision: Minister of Justice v Kim [2021] NZSC 57.

And on 13 April 2022 the Supreme Court, after receiving further information, assurances, and submissions, reinstated the Minister's decision to surrender Mr Kim for extradition: Minister of Justice v Kim [2022] NZSC 44.

Conviction appeals from judge-alone trials: review or rehearing?

On one of those subtle issues that would only occur to lawyers, the Supreme Court has decided how appeals under s 232(2)(b) of the Criminal Procedure Act 2011 against convictions in judge-alone trials, where the issue is the judge's assessment of the evidence, should be approached: Sena v Police [2019] NZSC 55.

They are not reviews, they are rehearings.

The difference? Reviews make allowance for the notion that reasonable minds may differ, and that, even if the appeal court might have reached a different conclusion, the judge’s verdict will be upheld if it was within the bounds of what was reasonable. Rehearings require the appeal court, once persuaded by the appellant that an error has occurred, to reach its own conclusion on the record of the evidence. But rehearings are not fresh hearings; the appeal court will recognise that in making findings of fact the trial judge had the advantages of seeing and hearing the witnesses, and had an overview which might not be available to appeal judges who may be given by counsel a rather selective view of the evidence. Where credibility is in issue, the appeal court will exercise “customary caution” (at [38]).

On rehearing, the appeal court will focus on whether it can be satisfied of guilt beyond reasonable doubt. This differs from a verdict insofar as if the court is not so satisfied it will usually order a new trial, rather than enter an acquittal, although the latter course may be taken in appropriate circumstances, such as long passage of time, the completion of a sentence, the unavailability of witnesses, or the compelling nature of fresh evidence of innocence.

Sena does not decide how appeals under s 232(2)(a) from jury verdicts should be approached. The Court was careful to specify that it was dealing with appeals from judge-alone verdicts.

Section 232 is silent on whether either of these sorts of appeals are reviews or rehearings. The Court’s extensive survey of the previous law and the various interpretations of earlier legislation demonstrates how the correct approach to these appeals has not always been clear.

There was some mention of appeals against verdicts in jury trials, under the old law. This was relevant because the first appeal in Sena had been to the High Court, which had applied the jury verdict approach to this judge-alone verdict (at [2]).

The old law on the reasonableness of jury verdicts had been considered by the Supreme Court in R v Owen [2007] NZSC 102, which I noted here. As summarised in Sena, under that law “the ultimate issue for the appellate court was whether the jury could not reasonably have been satisfied of guilt beyond reasonable doubt” (at [14]). This required exercising a review function, not the appeal court substituting its own view of the evidence for that of the jury.

So we are left wondering whether there will continue to be a difference between appeals from judge-alone verdicts (now, rehearings) and appeals from jury verdicts (used to be review, but what now?)

A clue, almost so subtle that if you look at it directly it disappears, is in footnote 43 of Sena.

No one is saying that the review-or-rehearing classification is watertight for all kinds of appeal. Appeals brought under s 232(2)(c), claiming a miscarriage of justice for any reason, might involve issues concerning the correctness of a trial judge's exercise of a discretion together with other issues that are appropriately considered by way of rehearing.

[Update: For application of Sena, (to a case decided before the judgment in Sena was delivered), illustrating a failure to give adequate reasons for convictions in a  judge-alone trial, see Webster v Police [2019] NZHC 1335.]