The Book of Why by Judea Pearl and Dana Mackenzie

Perhaps, after hearing rumours that artificial intelligence (AI) will replace judges and juries, you have picked up a book called The Book of Why – The New Science of Cause and Effect by Judea Pearl and Dana Mackenzie (2018). This sounds like just the thing to bring you up to speed on that topic.

The book is not really about replacing judges and juries, although it does deal with how to get computers to find causes by analysing data. Conventionally, unless a randomised controlled experiment has been conducted, data produced in an experiment says nothing about causes, only about correlations, although observational data, such as (for example) may be gathered in surveys and analysed by statisticians, can generate conditional probabilities from which causal effects are inferred. The main focus of the book is on how to avoid the need for randomised controlled experiments, and yet be able to say something about causes, or at least how to identify when there is no alternative to conducting a randomised controlled experiment to determine causal effects.

What is of potential interest to lawyers, it seems to me, is the diagrammatic representation of causal relationships that can arise in given circumstances. Can the kinds of diagrams described in this book give lawyers tools for analysis of issues in trials, where the fundamental question will be whether the defendant caused the relevant harm?

Here I have borrowed from Chapter 4 some of Pearl and Mackenzie’s examples of the use of causal diagrams. What I borrowed is the diagrams, and I have tried to apply them to the facts of cases I have imagined. Any mistakes are therefore mine, not the authors’. Decide for yourself whether this kind of approach may be of assistance in your work. There are, as the authors acknowledge, other ways of addressing whether cause has been established, but here is the causal diagram method, inasmuch as I currently understand it.

The authors claim the method is such fun that it can be thought of as a game.

The letters represent events or states of affairs (or, as one could say, variables, in which we may have varying degrees of belief), not people. I illustrate this in Game 1 for X and A, just so you get the hang of it.

Game 1

                                       

Defendant pulls trigger (X = this action by defendant)), gun fires (A = what the gun was caused to do), cartridge casing ejects (B), bullet hits victim (Y).

Here there is nothing to block the causal chain from X to Y. B is irrelevant to whether the firing of the gun caused the bullet to hit the victim.

Game 2

                                          

The prosecutor charges the defendant with participating in the manufacture of an illegal drug on two occasions, C and Y. The prosecutor's case is that the defendant had an appetite (A) for the illegal drug and that this appetite grew (X). On each occasion the appetite for the drug caused the defendant to collect (B and E) quantities of a substance that could be used to make the illicit drug, and his possession of this substance caused him to be accepted as a participant in the manufacturing process. However, the defendant explains (D) the accumulation of the substance by saying that the appetite is for that substance and that there was no purpose in accumulating it except to consume it.

Blocking (constriction) of a causal chain can occur if there is more than one possible cause for an event or a state of affairs. Here, there is a blockage in the prosecutor's causal chains A to C, X to Y, and B to E. B to E is blocked by the defendant’s explanation D. If the blockage is strong enough to maintain the causal direction D to B then C will be irrelevant to Y and A will be confined to enhancing X, which again will be irrelevant to Y because of the blocking of E. X to Y is blocked if D to E is of sufficient strength (ie if it carries sufficient credibility to raise a reasonable doubt about X to Y).

Game 3

                                      

                                           

This diagram can be taken in stages to reflect the course of a trial. The prosecutor alleges the defendant hated (X = the defendant's emotion) the victim, and this caused the defendant to kill (Y = what happened to) the victim. Hence the first step is the arrow from X to Y.

The defendant claims that it was another person who both (B) caused the defendant to hate the victim and killed the victim. There is now an arrow from B to Y. This creates a back-door cause, B to Y, of the victim’s death by the hand of that other person, which weakens the causal connection X to Y.

The prosecutor replies by adopting the defendant's allegation that the other person caused the defendant to hate the victim. Hence the arrow from B to X. The prosecutor argues that the causal chain B, X, Y strengthens the link X to Y.

The defendant adduces evidence of an alibi, that he is identified by security camera footage as being somewhere else at the time the victim was killed. This is represented by the arrow from X to A. The defendant hopes this will reduce the strength of the alleged causal link X to Y.

The prosecutor replies by challenging the identification of the defendant in the footage and says it really showed the other person who hated the victim. This is represented by the arrow from B to A.  

If the arrow B to A were to be reversed, because the defendant's alibi made it more likely that the other person who hated the victim was the murderer, there would be a causal chain A, B, Y, which the defendant would hope would reduce further the alleged link X to Y. The prosecutor's attack on the defendant's alibi was an attempt to block the suggestion that the other person was the murderer.

Game 4

                                         

Prosecutor alleges the defendant has large accumulated debts (X) which have caused the commission of the present fraud (Y), relying on propensity evidence: when the money problems began (A) they led to the defendant committing fraud (B), and contributing to that earlier fraud was a propensity for dishonesty (C), which also caused the presently alleged fraud (Y).

The proposed causal chain generated by the propensity evidence is blocked by C to B. This reflects the legal position that propensity alone is insufficient to prove current offending. There must be some evidence of current offending before the propensity evidence has anything to corroborate. If there is some such evidence, the admissibility of the propensity evidence will depend on the extent to which the prosecutor can reverse the causal link from C to B, for example by showing that previous commission of fraud (B) strengthened the defendant’s propensity for dishonesty. If the link then went from B to C, there would be no blockage in the causal chain X, A, B, C, Y.

Game 5

                                         

This is said to be a more realistic version of Game 4, with two new causal links B to X and X to Y. Using facts like those in Game 4, X to Y is the evidence necessary before propensity evidence has anything on which to operate (that is, to corroborate). B to X is the allegation that previous fraud (B) has caused present indebtedness (X), for example by resulting in reparation orders of a magnitude beyond the present means of the defendant.

Now the alleged causal link X to Y is supported by A, X, Y, and by A, B, X, Y, and by C, B, X, Y.

So, how secure is the position of judges and juries in coming to verdicts? Computers can't perform experiments, they can't use what the authors call "do-operators" to change things in the experimental environment. Where there are causal chains, data may be available to be used to generate conditional probabilities which AI can handle. But where a blocked chain cannot be ignored as being irrelevant, an experiment will be necessary to provide more data. A computer could not, therefore, play Game 2 because of the blocking effect of D (which would require an investigation of the likelihood that the defendant did consume the substance), or Game 3 because of B to A (unless it was able to compare the video image with the defendant, although there may be legal objections to that, or with the other person), or Game 4 because of C to B. Judges and juries surmount these difficulties with something ineffable called judgement.

Update: for more on AI and probabilities, including reference to this book by Pearl and Mackenzie, see the review by Paul Taylor in London Review of Books, 2021 vol 43, p37.

Go back!

What should be done about the failure of the Psychoactive Substances Act 2013 to control unapproved psychoactive products?

Unapproved psychoactive products should be controlled drugs. As such, their classification should follow recommendations of the Expert Advisory Committee on Drugs, and pending any advice to the contrary they should be Class C controlled drugs.

An arguably premature revision of the mental elements for liability for drug offences was undertaken in Cameron v R [2017] NZSC 89, [2018] 1 NZLR 161 (not freely available online, strange to say). [Background: perhaps the delay in its availability was due to outstanding proceedings against an absconding defendant keeping suppression orders in force until his death, reported on 2 March 2019 - although there had been extensive media coverage in 2018 of the entire case up to sentencing, with everyone named. No, that's not the reason. There is no reason.] The new requirement of knowledge that the substance was a controlled drug, or of recklessness as to its being a controlled drug, is too narrow in that it pinpoints “controlled drug”. This means knowledge of, or recklessness as to, the substance being scheduled in the Misuse of Drugs Act 1975, including knowing it by a name that is a common term for a scheduled substance (see Cameron at [40], [93]).

The problem here is that under this requirement it would be a defence for a person to have thought that the substance was an unapproved psychoactive product (there are currently no approved products, and none are in the process of being tested for approval).

Under the pre-Cameron law, the mental elements for guilt included knowledge that the substance was a drug that had an effect on the mind. This requirement, settled for some 30 years after some interesting cases were digested, did not give rise to intrinsic difficulties of proof. If anyone asked what I thought, I would say, go back to that. Cameron is suited to an anticipated future in which there are approved psychoactive products and it is necessary to recognise the innocence of people who honestly think they have such a product when in fact they have a controlled drug. On this view, we have here a rare example of reality not having caught up with the law.

As far as controlled drug analogues are concerned, we should wonder whether the legislated criterion of "substantially similar" structure to a controlled drug is really on point. The purpose of the Act is to prevent the harm that is caused by the use of controlled drugs, and substantial structural similarity serves (whether accurately or not, we don't necessarily know) as a proxy for that. In any event, for liability, mens rea should address the effects of the substance when used. If requiring knowledge of, or recklessness as to, structural similarity seems somewhat absurd (it has never been necessary for a defendant to know that a substance was an isomer, a salt, an ester or an ether of a scheduled substance) courts should be more comfortable assessing whether a defendant knew of, or was reckless as to, the effect of the substance on the mind of a user. People who manufacture and distribute such substances should be well aware of their effects when used, and of the similarity of those effects to those caused by use of controlled drugs. So, for offences in relation to analogues, there is no reason to treat the mental elements for liability as being any different from those applicable to other controlled drug offences. Our current legislation follows international precedents, at least as to definition of the actus reus, but that should not stop the search for improvements in the case law definition of the elements of mens rea.

It's a matter for you ... but you may well think ...

The High Court of Australia has accepted that, as a general rule, judges presiding at jury trials should not comment on issues of fact: McKell v The Queen [2019] HCA 5 at [31], [46]. Comment should only be used to correct an error or restore balance: [53]-[55]. The power to comment is to be used to ensure trial fairness, and is not to be used to add force to one side so as to sway the jury: [3].

It is important, obviously, that perceptions of trial fairness be kept consistent internationally. One reason for this is so as not to impede extradition processes.

The Supreme Court of New Zealand may well be alert for an opportunity to revisit the approach to judicial comment set out in R v Keremete CA247/03, 23 October 2003, applied recently in B v R [2018] NZCA 80.

Keremete tolerates a contradiction that is dispensed with by the general rule in McKell. This is that the judge may express in strong terms a view on the facts, while at the same time telling the jury that it is the sole arbiter of the facts. Keremete accepts a position in which strong comment can be consistent with fair presentation of the issues while at the same time leaving the issues of fact to the jury. Muddled, to say the least.

If jury trial judges disobey McKell, appellate courts will have to grapple with summings-up like that considered in B, where the court had to accept that there was prima facie lack of balance, but decided that in context the combined significance of the errors was considerably reduced, to the point where the court was able to conclude there was no unfairness.

This cleansing-by-context process is unsatisfactory, not the least because it is mysterious. B was not a case where comment was aimed at correcting an error or restoring an imbalance. It can only have been speculation for the court to conclude that there was no real risk that jurors had been influenced by the improper judicial comments. Trial outcome and trial process are separate matters, and it would be wrong to think that, because a verdict seems to have been correct, it was arrived at fairly. An analysis analogous (and here I stretch a bit for an analogy) to the requirements for the defence of withdrawal may be appropriate in determining whether a judge has restored the balance after making an inappropriate comment: compare Ahsin v R [2014] NZSC 153 at [140], applied in De Soto v R [2018] NZCA 366. Outside of strict criteria of that sort, breach of the McKell rule should, of itself, amount to a miscarriage of justice.

In what one hopes was a flash of brilliant wit, Gageler J declined to comment on judicial comments.

Fair process and choice of mode of trial

In S(SC 36/2018) v R [2018] NZSC 124, (the Court of Appeal decision is noted here on 5 May 2018) the defendant had not been told that a judge-alone trial was available, and he was convicted following trial by jury.

The Supreme Court delivered two judgments: William Young, O’Regan and Ellen France JJ jointly (delivered by Ellen France J, paras [1]-[86]), and a concurring judgment by Glazebrook and Arnold JJ (delivered by Glazebrook J, paras [87]-[99]). The appeal was dismissed.

The error was not jurisdictional ([46], [91]), for the court had jurisdiction to try the charges by either mode, and the error did not render the jury trial a nullity. It was a procedural error, and the issue on appeal was whether it had rendered the jury trial unfair. Was it unfair to try the defendant by jury, when he had not been aware of his right to a judge-alone trial?

The jury trial had been conducted fairly ([34], [96]). The defendant would have been advised to elect trial by jury ([54], [98]). There was no reason to think this was unsound advice ([56]). He had not demonstrated that he would have rejected that advice at the time it was given. On these facts there was nothing to suggest that an unfair procedural course had been taken ([83], [98]).

On different facts, if a defendant had been unaware of the right to elect trial by jury, that would amount to loss of an important statutory right, and such an error might, depending on the circumstances, amount to an unfair process rendering the judge-alone trial unfair ([99]).

So there could be circumstances where absence of advice on election of mode of trial would result in unfair process, but those did not exist in the present appeal ([96]).

There are some areas of vagueness. As far as nullity is concerned, a threshold is mentioned ([45]-[46]). This suggests some flexibility in the concept of nullity, and the cases cited indicate differing approaches and various circumstances. Nothing was resolved about that in this appeal, so the references to some of the other cases could be called excessive and perhaps even confusing. Cases interpreting different legislation need to be shown to be relevant, and older references to the overlapping of nullity and miscarriage of justice, and to degrees of nullity, do not seem relevant to s 232 in which attention is confined to three alternative ways in which a miscarriage of justice may occur: where there is a real risk that the outcome of the trial was affected, or where the trial was unfair, or where the trial was a nullity. Too much discussion of authorities calls to mind the amusing confession of having heard, “second hand”, that “a former clerk once referred to one of my judgments as ‘over-researched as usual’”: see Justice Susan Glazebrook, “Academics and the Supreme Court” (2017) 48 VUWLR 237 at 239. In this case it's the other judgment that seems rather overwrought.

Case Note: Onus and standard of proof that evidence was improperly obtained: Kearns v R [2017] NZCA 51, [2017] 2 NZLR 835

Criminal lawyers in New Zealand are familiar with s 30 of the Evidence Act 2006 which concerns the admissibility of improperly obtained evidence. Its first two subsections are as follows, and this note focuses on (2)(a):

“(1) This section applies to a criminal proceeding in which the prosecution offers or proposes to offer evidence if—

(a) the defendant or, if applicable, a co-defendant against whom the evidence is offered raises, on the basis of an evidential foundation, the issue of whether the evidence was improperly obtained and informs the prosecution of the grounds for raising the issue; or

(b) the Judge raises the issue of whether the evidence was improperly obtained and informs the prosecution of the grounds for raising the issue.

(2)The Judge must—

(a) find, on the balance of probabilities, whether or not the evidence was improperly obtained; and

(b) if the Judge finds that the evidence has been improperly obtained, 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 and takes proper account of the need for an effective and credible system of justice.”

On the issue of whether the evidence in question was improperly obtained, is there an onus on any party, and is there a standard to which the issue must be decided? The distinction between the terms onus and standard must be borne in mind. An onus, or as it is also called, a burden, is on the party that has an obligation to prove something. A standard is the level of proof or satisfaction required to be established by the party having the onus. On the ultimate issue of guilt in criminal cases, the onus or burden is (usually) on the prosecution and the standard is beyond reasonable doubt. Where admissibility is in issue, the decision may involve both determination of facts – a matter of proof – and consideration of those facts to decide whether impropriety existed – a matter requiring an evaluative assessment.[1]While there have been dicta[2]that suggest that it is not appropriate to apply a standard to the exercise of judgement, subsection (2)(a) seems plainly to apply the standard of the balance of probabilities to whether, as a matter of judgement, proved facts establish impropriety. Both the determination of the facts, and the evaluation of them, are subject to this standard.

Ordinarily, the party that asks the court to do something has the onus of satisfying the judge that it should be done. And ordinarily, technical legal terms have their usual meaning. So on first blush it would seem that the defendant (or, if applicable, a co-defendant) who asks the court to exclude evidence which is otherwise admissible, on the grounds that it was improperly obtained, has the onus of persuading the court that the evidence was improperly obtained. The standard is said to be on the balance of probabilities.[3]Even if it is the judge who raises the issue, the onus and standard should not change.

But in this context the phrase “on the balance of probabilities” could mean “on the preponderance of probabilities”. Here, preponderance means one scenario is more likely than the other, including where one reaches the balance of probabilities fixed standard, but also including cases where the more likely scenario is nevertheless below the balance of probabilities standard. This extended meaning of preponderance can be relevant because the judge is forced to make a decision. Whichever of the alternative scenarios is the most probable wins, even if both are less likely than not. In cases that may occur only relatively rarely there is another problem: what should the decision be if there is no reason to prefer one version of the facts to the other? Both versions may appear reliable, or they may seem equally unreliable. Here there is no preponderance of probabilities, and commentators have called this the equilibrium problem. A sensible solution to such a problem would be to refer to the onus of proof: equilibrium means the party with the onus has failed to persuade the judge to rule in its favour. So, does the absence of reference in s 30(2) to an onus mean there is no onus, or is an onus implied? And if an onus can be implied, is it always on one party, or can it shift depending on what the judge perceives as the justice of the case?

At an early stage commentators thought there was no onus, and that therefore equilibrium was a problem.[4] Priestley J recognised this in R v Hanford,[5]a case where the witnesses on each side seemed equally credible. The obligation to make a decision without an onus to resolve the equilibrium led his Honour to observe that this legislative framework infringed “the constitutional right of a judge, in rare cases, to be unsure”. If one respectfully doubts that there is such a constitutional right, preferring to say that litigants have a right to have judges decide their issues, it could still be said that a legislated requirement for a finding, in the absence of an onus, is an interference with the independence of the judiciary, even, in cases of equilibrium, to the extent of potentially requiring dishonesty. Certainly it is acceptable, and common, for the legislature to tell judges what matters to take into account, and even what matters are important, but what Priestley J found objectionable was the forcing of a judge to make a decision, in circumstances where a decision couldn’t be made.

In the light of this difficulty it would be worth re-considering the interpretation of these subsections. This is not to say that these issues matter in most cases, where decisions are relatively easy and outcomes would be the same regardless of onus, balance, and the imperative to decide. Many cases almost decide themselves, at least on the issue of impropriety, and it is not unusual for prosecutors to concede the point as a preliminary to contesting the balancing decision of admissibility.

In Kearns v R[6]the appeal against a ruling that evidence obtained in a vehicle search was admissible was allowed to the extent that the issue of admissibility was remitted to the District Court for further examination of the evidence and further findings of fact. This was necessary because the judge had failed to fully examine the lawfulness of the search and the extent of impropriety. The Court of Appeal’s mention of the issue of onus was therefore cursory, particularly because the Court was discussing the rather more interesting issue of the risk that the initial approach by the police was racially motivated. It was noted at [40] that, “Once a foundation is established, subs (2)(a) obliges the court, ideally with the assistance of counsel, to elicit the evidence necessary for a finding to be made.” At [37] the Court rejected a submission that it was for the prosecutor to prove that the search was lawful, once the issue was raised:

“The legislative history of s 30 demonstrates that Parliament specifically rejected imposing the burden of proving propriety on the prosecution. Parliament chose not to follow the Law Commission’s proposed draft in that respect ... . The correct approach is that neither party carries the onus. Onus of proof language does not make sense where the judge has a duty to make a positive finding. The judge must choose, as best he or she can on the evidence, which of the scenarios (proper or improper) is the most probable and make a finding accordingly. ....” [footnote omitted]

In the context of ordinary principles of interpretation, and of Hanford and similar cases, it is not controversial that the prosecution does not (always) have the onus of proving propriety. What stands out, however, is the claim that “Onus of proof language does not make sense where the judge has a duty to make a positive finding.” One would have thought, with respect, that an onus is particularly relevant where a finding has to be made, especially in those difficult cases of equilibrium. Perhaps the Court meant to say, “Standard of proof language does not make sense where the judge has a duty to make a positive finding”, which would be true but this revision does not fit the context. Be that as it may, the Court went on to emphasise that a choice must be made, as best as it can be, between the scenarios. One alternative must be more probable than the other, regardless of any standard. This seems to amount, in those difficult cases, to forcing a fiction of unequal likelihoods on the judge. Further, it requires the judge to conclude that evidence was improperly obtained just because that is the more probable scenario, even if the probabilities of each scenario are well below the standard of the balance of probabilities. The Court may have been taking the phrase in (2)(a), “on the balance of probabilities” to mean “on the preponderance of probabilities”.

It is respectfully suggested that the ordinary interpretation, mentioned earlier in this note, is to be preferred. It conforms with common sense as to onus, it applies the usual meaning of the balance of probabilities as a standard rather than a preponderance, it does not involve an equilibrium problem, it does not impose a fiction on judges, it does not compromise judicial independence, and it does not require highly unlikely allegations of impropriety to be accepted. If, however, a permanent onus on the defendant who seeks to have evidence excluded is thought to go too far, it can still be noted that the legislature’s refusal to place a universally applicable onus on the prosecutor, coupled with the silence of the provisions on the question of onus, does not exclude the use of an onus as the judge sees fit in the circumstances of a particular case.[7] 

[Update:] A shifting onus is consistent with the phrase in (2)(a), “whether or not”, which allows for the possibility that the prosecutor may succeed in satisfying the judge on the balance of probabilities that the evidence was not improperly obtained. The Law Commission has noted opinions to the effect that s 30 is biased towards holding improperly obtained evidence to be admissible, and has referred in this context to the equilibrium problem as illustrating that the party with the burden of proof (assumed to be the defendant) may fail to meet the standard of the balance of probabilities[8]. This appears to overlook the argument presented here.

---

[1] The phrase “evaluative assessment”, is used in Marwood v Commissioner of Police [2016] NZSC 139, [2017] 1 NZLR 260 at [46] to describe the process of assessing the appropriateness of a remedy for unreasonable searches, distinguishing it from the exercise of a discretion. Illustrations of the use of evaluation include assessing whether rights have been waived in the context of police interviews: Perry v R [2015] NZCA 530 at [32], the decision by a jury on the significance of propensity evidence: R(CA477/2014) v R [2015] NZCA 394at [23], [29], and the choice of an appropriate sentence: R v Cunnard [2014] NZCA 138 at [13]. An appeal against an evaluative assessment may involve the appellate court substituting its assessment for that of the court below. Where the basis for alleged impropriety is unfairness in the way the challenged evidence was obtained (s 30(5)(c)), an evaluative assessment will be required, as is illustrated on appeal in King v Police [2018] NZHC 1213 at [27]-[34] concerning a breach of the rules for taking statements, and in R v Alsford [2017] NZSC 42, [2017] 1 NZLR 710 at [38] as to breach of privacy.

[2] For example, R v Leitch [1998] 1 NZLR 420, 428, (1997) 15 CRNZ 321, 327 (CA) referred to in Simon France (ed) Adams on Criminal Law – Evidence (online looseleaf ed, Thomson Reuters) at ED2.02(2). The evaluation for impropriety can be straightforward, for example evidence obtained in an unlawful search will usually be obtained improperly, except where the unlawfulness is minor or technical: R v Williams [2007] NZCA 52, [2007] 3 NZLR 207, (2007) 23 CRNZ 1 at [21]-[23]. Often, a breach of the law can be established by proof of facts and without a further process of evaluation, but where reasonableness has to be assessed, for example in considering the adequacy of grounds for an action, evaluation will be necessary.

[3] The expression “the balance of probabilities” is discussed extensively in Z v Dental Complaints Assessment Authority [2008] NZSC 55. It is not a flexible standard (per Elias CJ at [28], Blanchard, Tipping and McGrath JJ at [101], Anderson J “largely in agreement” with the joint judgment [146]), and it means more likely than not (per Elias CJ at [28], joint judgment at [102]).

[4] 
     Mahoney et al, The Evidence Act 2006: Act and Analysis (2007, Brookers Ltd) at EV 30.06(1). The passage in Adams on Criminal Law – Evidence, above note 2, at EA30.06(1) has been to similar effect.

[5] R v Hanford HC Auckland CRI-2007-057-1922, 24 July 2008 at [8].

[6] 
     Kearns v R [2017] NZCA 51.

[7] Flexibility is apparent in pre-Kearns dicta. In R v Follas HC Rotorua CRI-2009-077-1516, 22 November 2010, at [37] Wylie J noted and agreed with Priestley J in Hanford at [82]-[84] (including the observation that s 30(2) is ambiguous on whether the defendant, or neither party, bears an onus), and considered that the onus should be on the prosecutor (at [38]). This was noted in, and is inconsistent with, Kearns (at [37]). It may be that in Kearns it would be easier for the prosecutor, than for the defendant, to establish why the police approached the car.

[8] 
     NZLC R142 The Second Review of the Evidence Act 2006 (28 February 2019), p 134 at [7.52] and footnote 43.

Reflections on Lundy

Good to see the Court of Appeal’s judgment in Lundy v R [2018] NZCA  410 (9 October 2018) delivered at last.

It’s one of those “world-famous-in-New Zealand” things: a double homicide that was always a whodunit. I don’t comment on the decision, because the case may not be over.

Some points do occur to me, however.

The appeal was determined under the old criteria for deciding conviction appeals (the new ones are here). Would the decision have been the same under the new law? Are the approaches in Bain and Matenga the same? See my notes of 11 May 2007  here and here, 9 July 2009 here, 20 July 2009 here.

To what extent would this illustration of the way conviction appeals are decided influence the future Criminal Cases Review Commission (CCRC) in its decision whether to refer cases to an appeal court? The current Bill proposing the establishment of the CCRC is here, and the referral criteria are in clause 17.

How appellate courts should approach conviction appeals is surprisingly unclear. Today the High Court of Australia corrected the Supreme Court of Western Australia on this: Rodi v Western Australia [2018] HCA 44 at [37], and, also today, the New Zealand Supreme Court granted leave to appeal on whether a conviction appeal after a judge-alone trial requires the appellate court to adopt a review approach or to substitute its view of the evidence for that of the trial judge: Sena v Police [2018] NZSC 92 (leave granted).

[Update: On 6 May 2019 the Supreme Court granted Mr Lundy leave to appeal: [2019] NZSC 45. Leave was limited to the proviso issue. We will have to wait to see whether there was an error made by the Court of Appeal insofar as it may have failed to consider the cumulative effect of the - admittedly, individually weak - inferences that could have supported the defence case. Could they, in combination, have raised a reasonable doubt? On this point we may recall how the reasoning of Ian Binnie in the Bain compensation case was criticised for (allegedly) not considering in combination points that supported the Crown's case. Place your bets. As any Bayesian bookie will tell us, if - for the sake of argument - a reasonable doubt exists when P(G) = 0.94 or less, then to move from a prosecution case that established guilt to P(G) = 0.99 to get to a reasonable doubt, you would have to accept defence evidence which, taken overall, was at least 6 times more consistent with innocence than with guilt. On the fairness side of the coin, broadly speaking, the issue would be whether there was a real risk that the wrongly admitted scientific evidence affected assessment of the probative value of the other evidence in a more than trivial way, by causing the jury to over-estimate the probative value of the evidence against the defendant or to under-estimate the probative value of the evidence that favoured the defendant.]

[Another update: after the Supreme Court delivered its judgment, I discuss the outcome here.]

When "must" a jury find a defendant guilty?

It is wrong for judges to compose, for use by juries in reaching verdicts, question trails that include directions that a verdict of guilty must be returned. For example, a question might take the form, “if [this question is answered “yes”] find the defendant guilty.”

Why is this wrong? A jury may be told when they must acquit, but never that they must convict. This is a safeguard against unconscionable prosecutions, where, although strictly speaking, the jury might be satisfied that the prosecutor has proved all the necessary elements to establish guilt, nevertheless it would be morally wrong to find the defendant guilty. The safeguard is so important that it is sometimes described as having constitutional importance. I have discussed this on 20 December 2013, here, and see also my note of 27 October 2006, here.

Yet it is wrong, at common law, to mention this option to the jury. Counsel cannot ask the jury to ignore the law. The law allows, but does not refer to, conscience verdicts. The subject belongs to jury-room discussion, into which a court will not inquire (as distinct from some aspects of jury-room discussion that might be the subject of judicial inquiry, for example to ascertain whether the jury has been improperly influenced).

Little appears to be known at the bar about conscience verdicts. [Update: see the references in S (SC 36/2018) v R [2018] NZSC 124 at [76], noted here on 22 December 2018.] For example, in De Soto v R [2018] NZCA 366, where the appeals were properly dismissed, no one raised the issue about the form that the question trail had taken. I do not suggest that any consequence followed from this in the circumstances of this case, as there does not appear to be anything about it that would raise a moral question about the appropriateness of findings of guilt. So the omission was not an error, and counsel may well have considered the point but correctly decided not to raise it in this particular appeal.

I just have a niggle with the question numbered 8 in [15] of De Soto: why is the instruction, in the event of an affirmative answer, to go on to question 9? The questions follow those set out in Ahsin v R [2014] NZSC 153 at [140], but they are put in the way designed to emphasise the Crown's burden and standard of proof. Ahsin requires a reasonable possibility that each of the following exist: demonstration of withdrawal, steps to undo encouragement, those steps being all that reasonably could have been done, and those reasonable steps having been taken in a timely way. Absence of a reasonable possibility that any one of these requirements has been met is sufficient for the defence to fail.

Also of interest in De Soto is the availability of the defence of withdrawal when a defendant is charged with importation of a drug, notwithstanding that the defendant may have been a principal offender through instigating the importation by placing an order with an overseas supplier.

Knowledge, belief, and recklessness in criminal law

Are the courts too willing to use an unsound definition of the mens rea requirement of knowledge?

A workable definition of knowledge is belief that is true. This has been used in cases and legal texts: for example, AP Simester and WJ Brookbanks, Principles of Criminal Law(4th ed, Brookers, Wellington, 2012) at [4.5]; David Ormerod, Smith and Hogan’s Criminal Law, (13th ed, Oxford University Press, Oxford, 2011) at [5.25]–[5.27], both cited in R v Banks [2014] NZHC 1244 at [39]. However, knowledge is more complex a concept than that, as discussed by Bertrand Russell in Chapter 13 of his The Problems of Philosophy (1912), which is freely available at http://www.gutenberg.org. A person does not really know something if the justification (that is, the reason) for the claimed knowledge, is false. Russell’s account of knowledge can be read with Edmund Gettier’s famous and brief 1963 paper “Is Justified True Belief Knowledge?”, available at http://fitelson.org/proseminar/gettier.pdf, which has generated a large body of entertaining discussion.

A true belief is not knowledge when it is deduced from a false belief. For example, the defendant may say, “I knew it was cocaine because it was some of what John gave me”, when in fact it was cocaine given to the defendant by someone else. The defendant believed that it was cocaine, on grounds that might have been strong if they were true. The reasonableness of the grounds depends on the likelihood of John having supplied the drug. If the grounds were weak, it would be more accurate to say that the defendant merely suspected that the drug was cocaine. This illustrates how a defendant may wrongly claim to have known something when it was only believed or suspected. True belief is also not knowledge if it is arrived at by fallacious reasoning. The defendant may claim to have known that a substance was cocaine because John supplies cocaine and John supplied this substance. The fallacy is that John may also supply other substances. Although the substance was indeed cocaine, the defendant’s belief would not normally be described as knowledge because it was a product of the mistaken assumption. The defendant had a belief, perhaps a belief on strong grounds if John nearly always supplied cocaine, but although the defendant thought it was knowledge, it was really just probable opinion. But if probable opinion is sufficiently probable to result in a firm belief, it can be correct to call it knowledge. This is different from the lesser probability required, for example, for the familiar “reasonable grounds to believe” needed in the search context, where an applicant for a warrant need only have the belief, not knowledge, that the relevant thing is at the place to be searched.

If a defendant admits in a police interview to having had knowledge of circumstances which were required to exist for there to be an actus reus, the police would be unlikely to challenge the defendant’s grounds for that claim of knowledge. But in the interval between interview and trial, when there would usually be some opportunity for philosophical reflection, a defendant may realise that the circumstances were more uncertain than they were thought to have been, and that the claim of knowledge was an exaggeration. This defendant may decide that what was perceived was a probability, not a certainty, and that the state of mind was suspicion rather than belief.

However, the law does tend to conflate the concepts of knowledge and belief. Whether it is correct to do so may be questioned. For example, see Kerr v R [2012] NZCA 121, which concerns knowing a purpose of future use of the relevant thing. The court, by interpreting knowledge as including belief (or, more accurately, as including belief that was false), in reality read in to s 12A(1) of the Misuse of Drugs Act 1975 the phrase “or believing”, referring to common law instances of knowledge being equated with belief (at [14]-[16]), the definition in the New Zealand Oxford Dictionary ([17]), and perceived difficulties with excluding belief ([18]-[19]). With respect, it should be noted that online Oxford dictionary definitions of  “know” do not refer to belief, and definitions of  “believe” do not refer to knowledge. Further, the Court’s difficulty with the concept of knowing a future event is beside the point, because what s 12A requires is knowledge of a present purpose (“is to be used”). The cases cited are from other contexts, but relevantly include where a substance was supplied to an undercover officer who had no intention of using it illicitly - raising the unaddressed issue of whether an investigatory technique should influence the definition of an offence - or receiving, where it was thought necessary to read in belief if knowledge was limited to what the defendant actually saw. This ignores the wider dictionary definition of know as to be aware of through observation, inquiry, or information. The real difficulty in Kerr was that Parliament had not anticipated that the police would use undercover officers to pretend to defendants that they were going to put items to illicit use, and the legislation should have specified "knowing or believing" in defining the mental elements of the offence.

Those matters aside, it may be that recklessness, which requires, as one of its elements, perception of a risk of the relevant criminality occurring if the defendant acts, is not amenable to the Russell - Gettier analysis. It is the perception, not the grounds for the perception, of the risk, that matters here. For the purposes of attributing criminal responsibility, where attention is on the harm threatened by the defendant, it is the perceived risk, not a known risk, that is a component of recklessness. The defendant’s perception of the probability of the relevant criminal outcome is sufficient, and if “knowledge” of the risk is referred to, it should be understood here in the loose sense of risk perception.

Take a coin, any coin

It is good to see an article on Bayesian reasoning with conditional probabilities in the current issue of the Times Literary Supplement: “Thomas Bayes and the crisis in science” by David Papineau (June 28, 2018).

As Professor Papineau points out, Bayesian analysis is used in many fields, including law.

One of the difficulties in discussing Bayesian reasoning, or indeed any complex subject, is that clear and simple points can become obscured by technical terms.

It took me a while to get to grips with Professor Papineau’s coin-tossing illustration. What it is designed to illustrate is an error of reasoning that is, apparently, found in too many published scientific studies. Essentially, the error involves drawing a conclusion from too little information. But Professor Papineau's illustration is of one test, not of an experiment involving a statistically significant number of test results.

If you take a coin – any coin – and toss it five times, and if you get five heads, how likely is it that the coin is biased? Pretend that you do not have special coin-tossing skills that allow you to determine the result of a toss. Also pretend that it doesn't occur to you to just keep tossing the coin to see what proportion of the sequences of five-tosses give results of five-heads.

After only a little reflection you realise that an unbiased coin will, on average, produce five-heads once every 32 times the five-toss sequence is carried out. One in 32 gives a probability of 0.03, approximately. The probability of getting five-heads from an unbiased coin looks very low, and you might be tempted to conclude that, therefore, there is a probability of 0.97 that the coin is biased.

Apparently, a significant number of scientific studies have been published in peer-reviewed journals, reporting conclusions arrived at through that sort of reasoning.

Bayesian analysis, if you are able to do it, will quickly show you that such conclusions are ridiculous, or, as Professor Papineau says, “silly” or “nonsense on stilts”.

If you are a lawyer, you might have to convince a judge or jury that an apparently obvious conclusion, reached by a respected expert, is wrong. It is far from easy to do this, and that may be why Bayesian analysis is taking so long to be routinely applied in courtrooms.

Fundamentally, the probability of getting five-heads if the coin is not biased, is not the same as the probability of the coin not being biased if it produced five-heads. The probability of A, given B, is not the same as the probability of B, given A.

My favourite way of illustrating this is to say: the probability of an animal having four legs, given that it is a sheep, is not the same as the probability of it being a sheep, given that it has four legs. The first tells you something about sheep, the second something about quadrupeds.

We know something about an unbiased coin: about three per cent of the times it is tossed five times it will produce a sequence of five-heads. But what do we know about a coin that has produced a five-head sequence? Is it biased or unbiased? If it is biased, how biased is it? Does it always produce five-heads or only some proportion of the times it is tossed five times? Is a biased coin commonly found or is it rare? Those things need to be known in calculating the probability that the tossed coin which produces a five-head sequence is biased.

At the risk of over-explaining this, let’s ignore - just for a moment - the rarity of biased coins and consider possible results of 100 five-toss sequences for a biased, and an unbiased, coin:

                                    Biased             Unbiased

            Five-heads       25                      3

             Other               75                     97

These results give three per cent of the results for the unbiased coin showing five-heads. The biased coin was, in this example, biased in such a way that it showed five-heads 25 per cent of the time and any other result 75 per cent of the time. So, of the five-heads results, three were from the unbiased coin and 25 from the biased coin, so the percentage of five-heads results that were from the biased coin is 25/28 times 100, or 89.3 per cent. Assuming you were equally likely to have tested either of the two coins, the probability, after one five-toss of the coin sequence, of the tossed coin being biased, given the five-head result, is approximately 0.89, which would not be regarded scientifically as significant proof of bias. Conventionally, for a significant conclusion that the coin was biased the conclusion could only be wrong no more than 5 per cent of the time.

This is not to say that the result is of no use. It does tend to prove the coin is biased. The strength of its tendency to prove bias is the likelihood ratio: the ratio of the probability of five-heads, given the coin is biased (from the above table this is 0.25) to the probability of five-heads, given the coin is unbiased (0.03), a ratio of 8.3 to 1. On the issue of bias, the result should be reported as: whatever the other evidence of bias may be, this result is 8.3 times more likely if the coin is biased than if it is not biased. The other evidence may be from a survey of coins which measured how often we can expect to find biased coins.

Now suppose that such a biased coin is only found once in every ten thousand coins, and that all biased coins have the same bias. The probability of a randomly chosen coin you have tossed and got a five-heads result being biased is, when you do the calculation using Bayes' formula, 0.0008. Eight occurrences in ten thousand. Much lower than the 0.97 probability (97 occurrences in 100) of the coin being biased that might have been reported in a peer-reviewed journal.

Again, this is not as surprising as it may seem at first glance. There may be only one biased coin in 10,000 coins, and one occurrence of five-heads from a biased coin in 40,000 coins (using the one-in-four frequency in the table), but, in round figures, there will also be 1200 occurrences (three per cent) of five-heads from unbiased coins in those 40,000 coins. This is why, on this occurrence of biased coins, a five-head result is much more likely (1200 times more likely) to be from an unbiased coin than from a biased one.

Only a very brave judge or juror would bet a significant sum that a coin which when tossed produced a five-head sequence was not biased. The bets would go the other way and those significant sums would most probably be lost.

And, as an afterthought: if you feel estimating prior probabilities is a bit haphazard, the Bayesian formula can be turned around to tell you what priors you would need in order to get in the above example P(the coin is biased) = 0.95. You would, before doing the experiment, need to be convinced to a probability of about 0.70 that the coin was biased. This sort of approach is discussed in a paper by David Colquhoun (available courtesy of The Royal Society Publishing). If, as a lawyer, you want an easy introduction to Bayesian reasoning, see my draft paper on propensity evidence.