Overwhelmed by prejudicial evidence

When is an error at trial insufficient to require an appeal court to quash a conviction?

The more the attention that is given to this question, the more difficult it is to answer. At least, that was the experience of appeal courts in the 2000s. Before then this question had gone largely unexamined, and appeals were decided with appellate courts deciding apparently intuitively whether miscarriages of justice were substantial enough to require the quashing of convictions.

I have considered this in discussing Weiss v The Queen [2005] HCA 81 on 16 January 2006, and see also the discussion of Grant v R (Jamaica) [2006] UKPC 2 on 20 January 2006. A difficulty was when should the appeal court act like a jury and consider the evidence and dismiss an appeal because despite the error at trial the evidence was sufficient to convict? See Bain v R (New Zealand) [2007] UKPC 33, discussed here on 11 May 2007.

The Weiss approach has been followed with minor adjustment in New Zealand: Matenga v R [2009] NZSC 18, discussed here on 9 July 2009.

An unattractive aspect of the Weiss approach is that it seemed to say that the appellate court may dismiss an appeal notwithstanding an error at trial if it considers that the evidence was sufficient to support a conviction. Obviously the ends (conviction of a person who is plainly guilty) cannot justify the means (never mind whether the trial was fair). To avoid that unpleasantness two requirements had to be met before a conviction could be upheld: the trial had to have been fair notwithstanding the error, and the evidence of guilt had to have been sufficient.

So, when does an error not affect the fairness of the trial? It is difficult to generalise without a definition of "fair trial". I tirelessly suggest one here.

Yesterday the High Court of Australia dealt with an appeal in which the errors at trial had been so substantial that it was unnecessary, indeed impossible, for the Court to address the sufficiency of the evidence: Patel v The Queen [2012] HCA 29 (24 August 2012). The appeal was allowed because no weight could be given to the verdict, the jury were overwhelmed by the prejudice created by the admission of evidence that, because of a change in the prosecution case, was inadmissible, and the proviso didn't apply: joint judgment of French CJ, Hayne, Kiefel and Bell JJ at [128]-[130]. Heydon J separately agreed, holding at [260] that there had been a departure from the requirements of a fair trial to such an extent that the court could not justly assess the strength of the case against the appellant, applying Gleeson CJ in Nudd v The Queen [2006] HCA 9, discussed here on 9 March 2006.

This seems to me to be a case where the unfairness was the rendering of the jury partial – that is, not impartial – through the introduction of evidence that would prejudice its judgment against the defendant.

Heydon J noted two difficult questions that did not have to be answered in this appeal [262]: to what extent should a judge disallow the introduction of evidence although no party objects to it, and when should a judge require the prosecution to provide particulars although they were not sought by the defence?

I suggest the answer to those is indicated by the meaning of a fair trial. But until judges get down to explaining what a fair trial means we are left to define it for ourselves.

I should add that in New Zealand we have revised the conviction appeal criterion by abolishing the proviso and defining miscarriage of justice to include errors that result in unfair trials: Criminal Procedure Act 2011, s 232 (not yet in force). Again, there is no definition of fair trial.

And now we are eight!

And that is 8 years of this site's existence.

It isn't really a blog in the sense of an invitation to readers to post comments. Instead it is an online casebook on criminal law. It now has commentary on over 500 cases from the leading appellate courts in New Zealand, Australia, Canada, the USA, the UK and the European Court of Human Rights.

My purpose is to promote scholarship and research in criminal law. For a given topic you can consult the Index which refers to the relevant cases and the dates on which commentary was posted. Or if you just want to know if there is commentary on a particular case, enter one word from its title in the search box on this page.

This site is not designed to make money, and there is no information here on how to contact me. It is just an attempt to show how interesting criminal law as a technical discipline can be. But it's not written anonymously because commentary needs to be responsible and readers should have an opportunity to assess the author's qualifications.

I have given the Blogger version of this site (http://www.nzcriminallaw.blogspot.com ) a title that now includes the word casebook; the mirror site retains its old title as I haven't figured out how to change it (http://www.donmathias.wordpress.com ).

When can one defendant rely on another defendant’s statement to the police?

If two defendants, A and B, are on trial together, when can A rely on part of B's statement to the police if B does not give evidence?

Broadly, once evidence is admissible it can be used for all purposes at trial unless there is a specific restriction: Hart v R [2010] NZSC 91 at [54]. Does s 27 of the Evidence Act 2006 provide for a restriction? Its first subsection says:

"Evidence offered by the prosecution in a criminal proceeding of a statement made by a defendant is admissible against that defendant, but not against a co-defendant in the proceeding."

Does this prevent the defence "offering" it? ("Offer evidence" is defined inclusively and widely in s 4.) Plainly no, although this has been controversial. Difficulties have concerned the meaning of "against". The answer seems to be that which is given in Leslie-Whitu v R HC Rotorua CRI-2009-263-163, 5 October 2011, Woolford J (not available online): s 27 does not prevent A relying on B's statement, but the statement is hearsay in relation to A. Since the evidence in relation to A must be considered separately from that in relation to B in a joint trial, A must rely on the hearsay provisions to get B's statement admissible in his case. As hearsay the rule is that it would be inadmissible, but there are exceptions, the main one of which is s 18 of the Evidence Act 2006.

The criterion in s 18 when B is unavailable, as he would be if he chose not to give evidence, is that "the circumstances relating to the statement provide reasonable assurance that the statement is reliable". The "circumstances" are defined in s 16.

I don't know whether that is the best answer. It creates an uneven playing field in that to use B's statement A must overcome the "reasonable assurance" of reliability threshold in s 18, whereas the usual requirement for admissibility is relevance, s 7, which has a low threshold: Wi v R [2009] NZSC 121, mentioned here on 11 August 2012. If the prosecution can get the statement admitted under s 27 (overcoming any objections based on unreliability, s 28, oppression, s 29, and impropriety, s 30) then A should be able to use it too without further restriction.

I suggested in a New Zealand Law Society Seminar that arguably the admissibility of B's statement for A's case should be governed by the common law fairness considerations.

Yesterday the High Court of Australia touched upon this in a common law context: Baker v The Queen [2012] HCA 27 (15 August 2012). However here B's statement did not really exculpate A: see Heydon J at [70]-[76]. But basically the Court said, "Aw fieck orf moit" (I assume Australians all sound like Hughsey).

The case illustrates how the common law has diverged in various jurisdictions. Canada and England (see Baker at [54], [88]) have expanded the hearsay exceptions to permit use of B's statement by A if it is apparently reliable, but as the judgments in Baker show, there are difficulties in establishing reliability in this context.

I still think use of the hearsay exception here is inappropriate because it is only by chance that B's statement is hearsay – at least under our statutory definition of hearsay which does not include within it statements made previously by a witness (see the definition of "hearsay statement" in s 4). If B gives evidence he is a witness so it is not hearsay, whereas if he does not give evidence it is. Should the trial tactics of one party govern the admissibility of evidence for another? And if the prosecution can rely on B's exculpation of A and argue it is a lie demonstrating B's guile, why can't A rely on it and argue it is a glimpse of B's truthfulness?

The common law position is different because the definition of hearsay is not dependent on whether the maker of the statement, B, gives evidence. Baker remains of interest for its emphasis on the difficulties in applying a criterion of reliability.

Scholarly and incite-ful

A small but fascinating point of general interest arises in The Queen v Khazaal [2012] HCA 26 (10 August 2012). If a defendant has the evidential burden of showing an innocent purpose, can evidence of general good character be sufficient to meet that burden?

The defendant in Khazaal compiled articles from other authors, edited them with his own introduction and published them under a pseudonym as an electronic book. The book contained advocacy of violence, but Mr Khazaal had a statutory defence if his publication was not intended to facilitate assistance in a terrorist act (see [48] and [51] of the joint judgment). The evidential burden here means the burden of adducing or pointing to evidence that suggests a reasonable possibility that his purpose was innocent.

Mr Khazaal suggested that his purpose was scholarly and peaceful, and he relied partly on the book itself to meet his evidential burden. The High Court unanimously held that the evidence relied on was not inconsistent with the alleged purpose of facilitating violence (French CJ at [19], Gummow, Crennan and Bell JJ at [77], Heydon J at [110]). Mr Khazaal also relied on his past pursuit of lawful activities as a journalist and argued that the book was written in that vein, however there was insufficient evidence to support that connection [111].

So, more generally, when does evidence of good character tend to prove a matter in issue? The New Zealand Supreme Court looked at this in Wi v R [2009] NZSC 121, holding that absence of previous convictions has sufficient probative force to meet the requirement of relevance (s 7 of the Evidence Act 2006) but in a particular case it may have only a slight tendency to prove a matter of consequence [19]. I mentioned Wi in the update to my note on Maye v R (Jamaica) [2008] UKPC 35.

Just a little bit of probative value is sufficient to meet the relevance requirement. This does not mean that almost anything goes, because the court can exclude evidence under s 8 if its prejudicial effect outweighs its probative value. Would that little bit of probative value be sufficient to meet the evidential burden of raising a defence? How much "tendency to prove" does the evidential burden require. Probably this is best expressed as a tendency sufficient to raise a reasonable possibility that the issue favours the defendant.

Here we drift further from the facts of Khazaal, like French CJ's ships passing in the night [12]. Imagine that 10% of scholars support violence, and 90% do not. Does evidence that D is a scholar mean that we can be 90% sure he does not support violence? Of course not. Relevance must not be confused with probative value. Yes, evidence that D is a scholar is relevant in the sense that D is using it to support the proposition that it tends to suggest he is not advocating violence. But the probative value (ignoring the separate question of rebuttal by evidence that D's actual writing did in fact advocate violence) of being a scholar requires consideration of different things.

First, one looks at the population of people who advocate violence, and asks what proportion of them are scholars. Then one looks at the population of people who do not advocate violence, and asks what proportion of them are scholars. This allows one to compare the proportion of scholars among the advocates of violence with the proportion of scholars among peaceful people. That tells one whether being a scholar is more consistent with advocating violence than it is with being peaceful. The probative value of the fact of D being a scholar is this comparison of proportions.

This is not how judges reason, but the result is the same in Khazaal. French CJ pointed out that mere consistency with innocent purpose is insufficient to tend to prove innocent purpose [18]-[20]. That is, scholars exist in both the population of advocates of violence and peaceful people, and their existence among peaceful people does not of itself have probative value for the proposition that D was peaceful. I suspect that he was wrong to adopt McClennan CJ at CL's phrase "the evidence ... was entirely neutral", because really the evidence was incomplete: it needed a context. The Court wanted more information about D's scholarly writing and about whether that was consistent with the writing which was the subject of the allegations.

It is appropriate to reject general probabilistic reasoning when specific evidence relevant to a particular case can be obtained. This is the Schrödinger's cat metaphor: once the box is opened and we look inside, the cat's being dead or alive is no longer a question of probabilities. If D kept the box closed, he needed to assist the court more with probabilities, and if he opened the box, he needed to demonstrate that this writing was in line with his other work.

Delay and its various remedies

The Privy Council has recently corrected any impression that may have existed about when unreasonable delay should result in the staying of proceedings or the quashing of a conviction: Tapper v DPP (Jamaica) [2012] UKPC 26 (17 July 2012).

The quashing of a conviction is only appropriate where delay prevents a fair trial or where delay makes it unfair to put the defendant on trial at all: Boolell v The State (Mauritius) [2006] UKPC 46 (noted here on 18 October 2006).

Other remedies may be the reduction of a sentence or if the defendant is acquitted the payment of compensation, the making of a public apology, or the expediting of a trial without further delay: Attorney-General's Reference No 2 of 2001 [2003] UKHL 68 per Lord Bingham at [24]-[25], applied in Tapper.

Issue estoppel and sentencing facts

A routine application of issue estoppel, applying R v Mahalingan [2008] SCC 63 (noted here on 15 November 2008) is R v Punko, 2012 SCC 39. In a multi-issue trial it could not be said that there was the necessary logical connection between the verdict and the issue now before the court, so there was no estoppel.

More interesting is the only point that divided the Court: Fish J disagreed with the majority judgment delivered by Deschamps J on whether a judge's finding of fact at sentencing could in principle give rise to issue estoppel. The majority held that it could not, because it was not a decision on the merits, that is, relevant to verdict, and issue estoppel requires a logically necessary conclusion that the jury were unanimous on the issue [11]-[12]. At sentencing the judge makes findings of fact to elucidate the jury's verdict of guilty, but in relation to counts on which the jury acquitted the defendant the judge in sentencing on the guilty verdicts has no power to make a finding binding on future judges [12].

Fish J, who agreed with everything else, considered that the sentencing powers of a judge included the power to find any relevant fact that was disclosed by the evidence, and that the question whether a fact found at sentencing could give rise to issue estoppel should be left open as a matter of principle [26]-[27].

One would have thought that if a sentencing judge finds a particular fact to have been proved beyond reasonable doubt, the issue is settled and the estoppel should apply.

Punko illustrates the narrowness of issue estoppel. Deschamps J did comment on how the doctrine of abuse of process could come into play if the prosecution's conduct were to be found to be "sufficiently egregious" [21]. In commenting on Mahalingan I noted that abuse of process can be of more use than issue estoppel in criminal law, if only the courts will get to grips with giving it meaning instead of dismissing it as something too vague.

Appellate error and wit

In the spirit of helpfulness ...

Don't rely on what our Court of Appeal said in Tutu v R [2012] NZCA 294 (5 July 2012) about there being a right to elect trial by jury on charges of common assault and assault on a constable in the execution of duty [19]. Section 43 of the Summary Offences Act 1981 removes that right.

For an insight into appellate advocacy and bench-bar exchanges, with occasional flashes of wit, in the High Court of Australia, see the perhaps rather irreverently-named http://shitjudgessay.tumblr.com .

Association with criminal groups

Some comments of general interest on criminal organisations or what are sometimes called organised criminal enterprises or organised criminal groups, were made by the Supreme Court of Canada in R v Venneri, 2012 SCC 33 (6 July 2012).

"[40] ... focus on the goal of the legislation, which is to identify and undermine groups of three or more persons that pose an elevated threat to society due to the ongoing and organized association of their members. All evidence relevant to this determination must be considered in applying the definition of "criminal organization" adopted by Parliament. Groups of individuals that operate on an ad hoc basis with little or no organization cannot be said to pose the type of increased risk contemplated by the regime.

"[41] Courts must not limit the scope of the provision to the stereotypical model of organized crime ― that is, to the highly sophisticated, hierarchical and monopolistic model. Some criminal entities that do not fit the conventional paradigm of organized crime may nonetheless, on account of their cohesiveness and endurance, pose the type of heightened threat contemplated by the legislative scheme."

Deciding whether there is a criminal organisation or an organised criminal group is a preliminary step. Liability for an offence depends on what conduct by a defendant is proscribed in relation to the organisation or group. Canada has defined several such offences, which broadly involve committing offences for the benefit of, or at the direction of, or in association with the organisation, or being a member of such an organisation and instructing any person to commit a qualifying offence for the benefit of the organisation. Also included are offences of participation in or contributing to the activity of the organisation with the purpose of enhancing the ability of the organisation to commit a relevant offence.

In New Zealand we have the offence of participation in an organised criminal group, if – again broadly - the defendant knows that his conduct contributes to achieving an objective of the group or to the occurrence of any criminal activity, or is reckless as to that contribution. The defendant need not share the objectives of the group.

It is not necessary that the organisation or group has actually committed any substantive offence, and contributing to achieving the objective of the group may not necessarily involve inciting or any other form of secondary liability.

Venneri highlights the point that associating with the organisation can occur through the defendant's offending and his membership is not required. So where the defendant supplied cocaine to a member of a criminal organisation, knowing that it was involved in drug trafficking, he was operating in association with the organisation, there was a sufficient nexus between the organisation and the defendant's supply of the drug.

Law reform by stealth

Sometimes the implications of law reforms or new legislation dawn on me rather slowly. This can happen to other people too, thank goodness, especially where controversial aspects of the proposed reforms distract attention from something more subtle but profoundly disturbing. So it is with ... [Note: the first part of this posting, down to the "Update" has been published in NZ Lawyer, Issue 188, 13 July 2012, copyright LexisNexis NZ Ltd - no longer available online as NZ Lawyer has a new publisher.]

Search on suspicion

The Search and Surveillance Act 2012, s 6 (not yet in force) will allow a search warrant to be issued on suspicion that a relevant offence has been committed, is being committed, or will be committed. This suspicion must be on reasonable grounds.

The reasonable grounds requirement looks like a safeguard. But the current law requires reasonable grounds for believing that a relevant offence has been committed or will be committed: s 198 of the Summary Proceedings Act 1957. The higher threshold of belief in s 198 is replaced by suspicion in s 6.

Certainly s 198 is not perfect. The new law improves it by referring to offences that are being committed. There is also a complexity in s 198 because it includes reasonable grounds for believing that a relevant offence is suspected of having been committed. But that has not been the subject of careful judicial analysis, and the case law has focused on the difference between belief and suspicion that an offence has been committed. In the leading case, R v Sanders [1994] 3 NZLR 450, (1994) 12 CRNZ 12 (CA) the universal requirement of s 198 was held to be belief on reasonable grounds both in relation to the offence and to the finding of the evidence.

Both s 6 and s 198 require reasonable grounds to believe that evidence of the relevant offending will be found at the place to be searched.

The Law Commission recommended that the threshold of reasonable grounds for belief should apply both to the offending and to the finding of evidence. In NZLC R97 (2007) "Search and Surveillance Powers" the Commission said (Recommendation 3.1):

"There should be a standard statutory threshold for the exercise of general law enforcement powers of search. That threshold should be reasonable grounds to believe that an offence has been, is being, or is about to be committed, and that evidential material is in the place to be searched. That test should be departed from only where there is a compelling case to do so. Current provisions where the lower threshold can be justified and should be retained include sections 60 to 61 of the Arms Act 1983 and search powers relating to border control offences under the Customs and Excise Act 1996."

The Search and Surveillance Bill was so extensive and contained so many controversial measures that submissions do not seem to have focused on what has become s 6. I didn't notice it myself. Either everyone thought it was alright, or everyone overlooked it. There is no mention of it in the minority views in the Justice and Electoral Committee's report on the Bill.

The courts certainly think the distinction between reasonable grounds to believe and reasonable grounds to suspect is significant. See Collins v R [2010] NZSC 3 at [2]. In Britten v R [2012] NZCA 81 at [15], Priestley J for the Court cited R v Williams [2007] NZCA 52, [2007] 3 NZLR 207 at [213]:

"Having "reasonable grounds to believe", the test under s 198 of the Summary Proceedings Act, is a higher standard to meet than "reasonable ground to suspect", the test under s 60(1) of the Arms Act for example (see R v Karalus (2005) 21 CRNZ 728 (CA) at para [27]). 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 (see Laugalis (1993) 10 CRNZ 350 (CA) at pp 354 – 355), 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 (see Sanders [1994] 3 NZLR 450
at p 461)."

The distinction is important too where there are powers of search without warrant. For example s 18(2) of the Misuse of Drugs Act 1975 permits warrantless search where there are reasonable grounds for believing in both the occurrence of a relevant offence and the presence of evidence. However s 20 of the Search and Surveillance Act 2012 replaces that with a power to search without warrant where there are reasonable grounds to suspect the commission of a relevant offence. (I note in passing that it is strange, given the opportunity to improve the law, that s 20 retains the rather technical and unrealistic requirements of s 18(2) as to identity of drug, particularly in view of the difficulty that they had caused the prosecution in Hill v Attorney-General (1990) 6 CRNZ 219 (CA)). The significance of the high threshold for warrantless search was emphasised in Hill by Richardson J at p 222:

"Search without warrant is a very special exception to the general principle that police officers must satisfy a statutory officer that they have reasonable grounds for believing justifying the issue of a warrant authorising their entry to and search of private property. As Sir Thaddeus McCarthy put it, "New Zealanders have been cautious in our grants of powers to the police", citing Lord Devlin's observation that:

"It is because the British have learnt to measure out stingily their grants of authority so that it is just enough and no more that they have, perhaps more successfully than any other nation, held the balance between order and freedom"

("The role of the Police in the Administration of Justice", Clark, Essays on Criminal Law in New Zealand, p 173.)"

We now seem to be intent upon eroding this general principle of requiring a belief, and a reasonable belief at that. Does our legislature's eagerness to grant powers to the police really reflect New Zealanders' wishes? Recent high profile searches have arguably given fresh life to Sir Thaddeus's reflections. We should reconsider this issue.

[Update] After I posted the above comments a colleague drew my attention to a recently published New Zealand Law Society Seminar Paper on the Search and Surveillance Act. There, the authors review the changes made to the grounds for issuing search warrants, in particular under s 6, and they conclude that the provision is balanced:

"Law enforcement needs are met by not requiring an overly high factual foundation for the commission of a crime to trigger investigative powers. But human rights, and in particular the right to be free from unreasonable search (and by extension having reasonable expectations of privacy respected) are upheld by requiring a more significant factual foundation for those investigative powers to operate on a particular place or person, etc."

It doesn't surprise me that this was written by two prosecutors. They wrongly suggest that the belief as to the existence of the evidence is a protection of human rights and they overlook the erosion of the same rights created by the "[not] overly high factual foundation" of suspicion. It is easy to see that the belief in the presence of the evidence at a specified location is readily established: for example, if a reliable witness saw items being taken into a stated address, the reasonable grounds to believe they were there would be made out. That would say nothing about the requirement of the items being evidence of an offence, which is, under s 6, easily met by other information that gives rise to a suspicion of criminal activity. It is this element that protects - or should protect - reasonable privacy expectations.

It is fair to expect confirmation of my assertion that reasonable grounds to believe that a specified offence has been committed are required by s 198. In Rural Timber Ltd v Hughes [1989] 3 NZLR 178 (CA) the Court included in its description of the effect of s 198 the following:

"The application and warrant should specify which of those lettered paragraphs [(a), (b) and (c) of subsection (1)] is involved. Generally speaking as many of them as it is proper to invoke should be invoked and a proper foundation should be laid for the necessary reasonable grounds for belief in respect of each paragraph invoked. In almost all cases where the issue of the warrant is justified paragraph (b) is likely to apply; paragraph (a) is generally somewhat less likely to be engaged while paragraph (c) is probably the least likely to apply in the general run of cases"

And the form prescribed for search warrants, Form 50 in Schedule 1 to the Summary Proceedings Act 1957 materially as to paragraph (b) states:

"To every constable ... I am satisfied ... that there is reasonable ground for believing that there is in [the specified place, specified things] ... which there is reasonable ground to believe will be evidence as to the commission of an offence of [specified] ... ."

This applies the standard of reasonable grounds to believe to both the presence of the evidence in the specified place and to the link between it and the commission of a relevant offence.

A search authorised on lesser grounds, such as suspicion that a relevant offence has been committed and that the thing searched for will be evidence of that offence, will be a limitation on the right to be free from unreasonable search: s 21 New Zealand Bill of Rights Act 1990 (BORA).

Yet, on 12 June 2009 the Crown Law Office advised the Attorney-General that the search powers in what was then the Search and Surveillance Bill (now enacted, including s 6) did not give rise to unreasonable search and seizure in terms of s 21 of BORA. That ignores the restriction of the meaning of unreasonable search that is required for the new law to be accommodated. Unreasonable search will no longer include searches based on grounds that only amount to suspicion. The advice did not specifically address the terms of what is now s 6.

The Law Commission advised the Justice and Electoral Committee, which reported to Parliament on the Search and Surveillance Bill, that what is now s 6 does not substantively change the grounds for issuing a search warrant. (See the Interim Report on the Search and Surveillance Bill 45-1, p 10.) But s 6 is inconsistent with the Law Commission's Recommendation 3.1 quoted above which in turn is consistent with s 198. Plainly the point was overlooked, as there was so much in the Bill requiring consideration.

What are the mental elements of dangerous driving?

Dangerous driving is risk-based, not negligence-based: King v The Queen [2012] HCA 24 (20 June 2012). It is not necessary to ask whether there has been a sufficiently great level of negligence to attract criminal rather than merely civil liability. Such negligence could be sufficient if it reached a high enough level, but it is not a necessary ingredient of dangerous driving. So, on the 3-2 majority's reasoning in King, the judge had not been wrong to direct the jury that the prosecution did not have to prove criminal negligence [48], [50].


The appellant submitted that the judge had steered the jury away from what in the context of the trial was conviction on the lesser offences. More serious than dangerous driving causing death were the offences on which he had been convicted, culpable driving causing death. The judge, said the appellant, had given the jury the impression that dangerous driving causing death was too trivial an offence to reflect his culpability, as it didn't even require criminal negligence. In diminishing the seriousness of dangerous driving (here, the conviction sought by the defendant) the judge may have conveyed to the jury that the more serious offence of culpable driving could more easily be committed.


You can see that while the judge had "not been wrong" to direct the jury that criminal negligence did not have to be proved for dangerous driving, at the same time a direction that criminal negligence was required would also have been correct in the circumstances. Such negligence would be sufficient, albeit not necessary for liability. The majority should have held that the judge was wrong in the context of the evidence. The minority, Heydon and Bell JJ, would have allowed the appeal on the grounds that this had been a misdirection which had created a substantial miscarriage of justice because the defendant may have been deprived of a more favourable verdict. They substantially agreed with the majority on the elements of dangerous driving.


When a court asks whether the driving in question demonstrated a marked departure from the standard of care that a reasonable person would observe in all the circumstances (R v Roy, 2012 SCC 26, discussed here on 15 June 2012), it is focusing on one sort of dangerousness: that which is accompanied by negligence. Usually examples of dangerous driving will involve some degree of negligence, above a minimum level (marked departure from the reasonable person's standard of care), and this can include even momentary inattention in circumstances where particular care and attention are required (King at [46]). However negligence is not an element of dangerous driving (Bell J [94]), just as (this is me now, not Bell J) stabbing is not an element of murder. But just as in the circumstances of a given case stabbing may have to be proved to secure a conviction for murder, so too negligence may have to be proved to secure a conviction for dangerous driving.


It is inevitable that jurors will have to assess the driving in the instant case against an imagined scale of qualities of driving. Somewhere on the scale careless driving becomes dangerous driving, and somewhere else on the scale dangerous driving becomes a more serious offence. The statutory context is always important. The extent to which the defendant departed from the objective standard required by the law for all drivers has to be assessed (Bell J at [74]). It is of no use to tell jurors that the defendant's driving must be such as to "merit criminal punishment" (Bell J [74], Heydon J [68]) because the criterion for guilt should not be left to jurors but rather is for the law to lay down. The criterion is whether the manner of driving gave rise to potential danger in a real sense to a person in the vicinity (French CJ, Crennan and Kiefel JJ at [46], applying Barwick CJ in McBride v The Queen [1966] HCA 22, (1966) 115 CLR 44). Plainly, in assessing responsibility for dangerous driving it is necessary to compare the level of danger to that required for liability for comparable offences. This does not mean that comparison with civil law negligence is useful: Bell J at [102].


Bell J stated [92] the mens rea requirement for dangerous driving as being confined to "the intention to do the acts involved in driving". She said that liability is subject to an objective test, citing the old case R v Gosney [1971] 2 QB 674 at 679. But in Gosney D 's conviction was quashed because she had made the same mistake as a reasonable driver would have made, and the Court of Appeal held that dangerous driving was not an offence of strict liability (for this point I rely on my 1978 Fourth Edition of Smith and Hogan, Criminal Law, pp 88-89). Bell J's dictum on mens rea is difficult to reconcile with negligence being a sufficient but not a necessary condition for liability.


By comparison, the Supreme Court of Canada required in Roy, above, mens rea elements of either an unreasonable failure to recognise a risk of the required kind or an awareness of the actual risk created by the driving and a deliberate running of that risk. The culpability or fault requirement inherent in the notion of mens rea is generally said to consist of either intention or recklessness as to the essential factual elements of an offence. But the meaning of intention in criminal law is context-dependent, sometimes extending to consequences and sometimes not.


Since we are not allowed to include criminally negligent failure to recognise danger as an element of the offence, we must slip it in as a relevant consequence under intention. The mens rea elements of dangerous driving here would therefore be: either intention to drive with knowledge of the danger created by the manner of driving, or intention to drive accompanied by a failure to appreciate the danger that a reasonable person would recognise as being created by the manner of driving.

Responsibility for dangerous driving

The offence of dangerous driving is carefully analysed in R v Roy, 2012 SCC 26 (1 June 2012), where the appellant had been convicted of dangerous driving causing death. He had pulled out from a stop sign into the path of an oncoming vehicle. There was thickening fog and visibility was poor.
Cromwell J, for the Court, observed [1]:

"Dangerous driving causing death is a serious criminal offence punishable by up to 14 years in prison. Like all criminal offences, it consists of two components: prohibited conduct — operating a motor vehicle in a dangerous manner resulting in death — and a required degree of fault — a marked departure from the standard of care that a reasonable person would observe in all the circumstances. The fault component is critical, as it ensures that criminal punishment is only imposed on those deserving the stigma of a criminal conviction. While a mere departure from the standard of care justifies imposing civil liability, only a marked departure justifies the fault requirement for this serious criminal offence."

On the reach of the criminal law, he continued:

"[2] Defining and applying this fault element is important, but also challenging, given the inherently dangerous nature of driving. Even simple carelessness may result in tragic consequences which may tempt judges and juries to unduly extend the reach of the criminal law to those responsible. Yet, as the Court put in R. v. Beatty, 2008 SCC 5 (CanLII), 2008 SCC 5, [2008] 1 S.C.R. 49, at para. 34, "If every departure from the civil norm is to be criminalized, regardless of the degree, we risk casting the net too widely and branding as criminals persons who are in reality not morally blameworthy." Giving careful attention to the fault element of the offence is essential if we are to avoid making criminals out of the merely careless."

On 25 February 2008 I discussed Beatty and its various judicial analyses of the ingredients of dangerous driving. A minority approach in that case was to use mens rea to assist in determining whether there was an actus reus, which indeed seems conventional (actus non facit reum nisi mens sit rea, or words to that effect), but it ignores the first requirement which is an act that causes harm. A level of bad conduct must exist before it can come within the reach of the criminal law. Otherwise, outwardly normal driving would be criminally dangerous if the driver's state of mind was a marked departure from the standard of care expected of drivers. In Beatty all judges agreed that the acquittal should be restored, but their reasoning was divergent. McLachlin CJ, Binnie and LeBel JJ held that a momentary crossing of the centre line was not an actus reus of dangerous driving. Charron, Bastarache, Deschamps, Abella and Rothstein JJ held – correctly I think - that it was. All judges agreed there was no mens rea as there was only a momentary lapse of attention.
So the Court's attention in Roy was on the sensitive issue of actus reus. It is not, said the Court, the consequences (collision and death) that determine whether there was an actus reus, but the quality of the driving itself [34]:

"In considering whether the actus reus has been established, the question is whether the driving, viewed objectively, was dangerous to the public in all of the circumstances. The focus of this inquiry must be on the risks created by the accused's manner of driving, not the consequences, such as an accident in which he or she was involved. As Charron J. put it, at para. 46 of Beatty, "The court must not leap to its conclusion about the manner of driving based on the consequence. There must be a meaningful inquiry into the manner of driving" (emphasis added). A manner of driving can rightly be qualified as dangerous when it endangers the public. It is the risk of damage or injury created by the manner of driving that is relevant, not the consequences of a subsequent accident. In conducting this inquiry into the manner of driving, it must be borne in mind that driving is an inherently dangerous activity, but one that is both legal and of social value (Beatty, at paras. 31 and 34). Accidents caused by these inherent risks materializing should generally not result in criminal convictions."

Personally I would take the consequences of the driving as relevant to whether there was an actus reus: what was the harm? Not just pulling out from a stop sign. Hitting another vehicle was relevant too. Be that as it may, the Court continued by reasoning that if there is an actus reus, attention turns to whether there is a fault element or mens rea:

"[36] The focus of the mens rea analysis is on whether the dangerous manner of driving was the result of a marked departure from the standard of care which a reasonable person would have exercised in the same circumstances (Beatty, at para. 48). It is helpful to approach the issue by asking two questions. The first is whether, in light of all of the relevant evidence, a reasonable person would have foreseen the risk and taken steps to avoid it if possible. If so, the second question is whether the accused's failure to foresee the risk and take steps to avoid it, if possible, was a marked departure from the standard of care expected of a reasonable person in the accused's circumstances."

The "marked departure" requirement serves to separate criminal from civil responsibility. Momentary lapses of attention can occur in good drivers and in appropriate circumstances may give rise to civil liability or to convictions for careless driving [37]. As established in Beatty and cited by the Court in Roy [38]:

"The marked departure from the standard expected of a reasonable person in the same circumstances — a modified objective standard — is the minimum fault requirement. The modified objective standard means that, while the reasonable person is placed in the accused's circumstances, evidence of the accused's personal attributes (such as age, experience and education) is irrelevant unless it goes to the accused's incapacity to appreciate or to avoid the risk (para. 40). Of course, proof of subjective mens rea — that is, deliberately dangerous driving — would support a conviction for dangerous driving, but proof of that is not required (Charron J., at para. 47; see also McLachlin C.J., at paras. 74-75, and Fish J., at para. 86)."

And now for the tricky point, contentious in Beatty: when can an inference of mens rea be drawn from the actus reus? The Court gave this answer [42]:

"Driving which, objectively viewed, is simply dangerous, will not on its own support the inference that the accused departed markedly from the standard of care of a reasonable person in the circumstances (Charron J., at para. 49; see also McLachlin C.J., at para. 66, and Fish J., at para. 88). In other words, proof of the actus reus of the offence, without more, does not support a reasonable inference that the required fault element was present. Only driving that constitutes a marked departure from the norm may reasonably support that inference."

The trouble with this approach is that it is perfectly normal for the law to hold a person responsible for an intended act that causes harm, and intention is usually inferred from the conscious performance of an act. There is no such thing as driving that is merely "simply dangerous" because by definition dangerous driving is driving that is a marked departure from the standard of a reasonable driver. This imagined "simply dangerous" driving should not be immune from the usual approach to determining intention and responsibility, and "on its own" it should support an inference that it was intended. The Court acknowledged [42, above] that "simply dangerous" driving is an actus reus. It is important to remember we are only talking about the drawing of an inference of mens rea, and in the circumstances of a particular case there may be facts giving rise to competing inferences or facts that strengthen the inference of mens rea.
The Court in Roy was concerned to avoid extending the reach of the criminal law beyond what was appropriate in view of the law's purposes. It appropriately set a high level for the required harm (the actus reus) but then needlessly constrained the process of inferring the existence of the mental requirements for criminal responsibility. In Roy the trial judge had inferred mens rea from the fact that the driving was objectively dangerous, and the Supreme Court held that that was wrong [44]. Given the thickening fog, the stop sign, the poor visibility and the oncoming traffic, the Court concluded [51]:

" ... While the appellant's act of driving out from the stop sign was apparently a voluntary act, there was no evidence to support the conclusion that the appellant was in fact aware of the risk he was creating in doing so and deliberately chose to run that risk ... ."

This seems to mean that the evidence suggested that the appellant had been well aware of the risk but had simply, and excusably, misjudged the position and speed of the oncoming vehicle. He knew there was a risk but he did not correctly measure the level of risk. The Court's assessment of the facts was [55]:

" ... on any realistic scenario consistent with the evidence, the time between visibility and impact would be only a few seconds. In my view, the appellant's decision to pull onto the highway is consistent with simple misjudgment of speed and distance in difficult conditions and poor visibility. The record here discloses a single and momentary error in judgment with tragic consequences. It does not support a reasonable inference that the appellant displayed a marked departure from the standard of care expected of a reasonable person in the same circumstances so as to justify conviction for the serious criminal offence of dangerous driving causing death."

Yes indeed, the defendant was not guilty because of absence of mens rea. I think there was an actus reus, just as there really was one in Beatty: pulling out from a stop sign and hitting an oncoming vehicle is a dangerous act, just as crossing the centre line while driving and hitting a car on the other side of the road is a dangerous act, but in each case the absence of mens rea supported acquittal. There was not the degree of inattention required in the circumstances to support a finding of sufficient fault for attribution of criminal responsibility.

Criminal responsibility, insanity, diminished responsibility and the role of the common law

The criteria for holding a person (D) responsible in criminal law for a particular harm emerge from moral decisions. These criteria are manifest as statutory laws and as judicially created common laws. For example, D's acts must be voluntary. They must cause the relevant harm. There must be no circumstance of justification or excuse.

The recent Supreme Court of Canada case Maybin v R, 2010 SCC 24 noted here, highlighted the moral basis for attribution of a sufficient causal connection between D's acts and the harm. The Privy Council has recently referred to the moral dimension of diminished responsibility: Daniel v The State (Trinidad and Tobago) [2012] UKPC 15 at [37]-[38].

The moral dimension is often recognisable by a requirement for reasonableness. Self-defence justifies killing, or any other form of violence, where it is reasonable for D to cause the harm in the circumstances as he believed them to be. This almost universal formulation of self-defence reflects its common law origins. The common law reflects the customs of the people. Reasonable conduct should be lawful because it is perceived as right.

Is the common law merely residual, ever-diminishing as statute advances? Or is the common law a potentially powerful source of justice? It might seem from the High Court of Australia decision in PGA v The Queen, discussed here recently, that the common law perpetually withers, although that would  - or should - be a misconception. Judges deal with the requirements of justice in particular cases, and with each case there is potential for development of the common law. This should be as true for defences as it is for any other developments that justice requires. By "justice" I mean here the judicially perceived sense of what the community regards as right. For recognition of this ability to develop new defences, see R v Kingston[1994] 3 WLR 519, at 536 (although in R v Cargill [1995] 3 NZLR 263, (1995) 13 CRNZ 291(CA) I was unable to persuade the Court to recognise a new defence of reasonable demand in relation to a charge of blackmail; that development was subsequently made by legislation: Crimes Act 1961, s 237(2)).

Parliament tends to limit defences and to create new offences. It seems to be engaged in extending the scope of acts for which a person can be held criminally responsible. Whether that legislative activity really reflects community perceptions of what is right can be a controversial question. But the important point is that if the courts are perceived by Parliament to be going wrong, legislation can correct that error. Whatever the common law might say, it is trumped by legislation. This is why courts should not be shy about developing the common law.

Daniel, above, prompts us to think about whether a defence of diminished responsibility should be developed at common law where legislation is silent on that. A defence of insanity will not cover all the states of mind that should negate criminal responsibility. The varieties of what in a loose sense may be called mental disorder seem extensive, as anyone who has glanced at the Diagnostic and Statistical Manual (4th ed) can see. Some of them, as Daniel illustrates, are states of mind that, if established at trial, should lead to acquittal. A court should be allowed to decide whether in a particular case mental abnormality should attract liability. This is what the court in Norway is currently doing, but under the guise of deciding the question of sanity.

An overview of judicial response to mental deficiencies in the context of the (now, former) statutory partial defence of provocation in New Zealand illustrates how moves from absolute rejection to accommodation can be made. In R v McGregor [1961] NZLR 1069 (CA) the Court declined to extend the meaning of provocation in the absence of statutory authority. Some withdrawal from this absolute position is evident in R v McCarthy [1992] 2 NZLR 550, (1992) 8 CRNZ 58 (CA). A variety of mental weaknesses have been accepted as relevant to whether D had in a non-culpable sense lost his self-control: R v Taaka [1982] 2 NZLR 198 (CA) obsessive compulsive personality, R v Pita (1989) 4 CRNZ 660 (CA) aversion to violence, R v Foreman 18/2/92, CA254/91 active schizophrenic process, R v Oakes [1995] 2 NZLR 673 (CA) battered woman's syndrome, R v Timoti [2005] 1 NZLR 466, (2004) 21 CRNZ 90 (CA) paranoid personality disorder.

Does abolition of provocation as a partial defence to murder mean that none of these factors can have any relevance to the attribution of criminal responsibility generally? I think not. But even on a narrow view that they could only be relevant to whether intention or recklessness, or voluntariness, or causation had been proved, there is sufficient flexibility in the mysteries of fact-finding for morality to have an influence.

Ascertainable law

The law must be ascertainable. This is a requirement for the enforceability of enacted law and for the validity of common law. In jurisdictions where the common law creates offences, it must be possible to find out what they are.

It doesn't matter how sure we now are of what the common law used to be, if it was uncertain when a person did acts that are now alleged to have been contrary to that law.

In PGA v The Queen [2012] HCA 21 (30 May 2012) the issue was whether in 1963 in South Australia, where rape was a common law crime, a wife could withdraw her consent to sexual intercourse with her husband.

To find that out, one would think that it would be sufficient to consult the case law of the time. Were husbands charged with raping their wives? Did judges say that that was rape? What did legal scholars think? If a man went to his lawyer then and asked, what would he have been told?

According to the majority in PGA, a lawyer back then would have concluded that a wife could withdraw her consent to sexual intercourse with her husband. This view would have been reached by considering the fact that marriage is not permanent [59], that marriage is a consensual contract which could be breached by wilful and persistent refusal of sexual relations – implying that some refusal is possible [60], that relevant legislation required cohabitation but did not require matrimonial intercourse as an order to that effect would be unenforceable [60], that a wife had a distinct legal personality from her husband, particularly as to property rights, and this undermined any idea of her being her husband's property [61], and that women had the vote and had recognised rights to commercial and professional lives [60]-[61].

These considerations, according to the majority reasoning, would have led a lawyer in 1963 to advise that a husband could be guilty of raping his wife.

It is for judges or the legislature to change the common law. The majority review at [19]-[24] the meaning of the term "the common law", emphasising how judicial precedents may diminish in value over time. Old decisions can become obsolete and inoperative.

But, one would have thought, it is for judges to say when that obsolescence and inoperativeness has led to a change in the common law, and to declare the new law for the future. The majority in PGA have made that declaration retrospectively. One does not have to say that they are wrong in deciding that a judge in 1963 would have made the same declaration, to recognise the injustice from the point of view of the defendant who now faces historical charges including two counts of rape of his then wife.

Our natural aversion to rape may cloud perception of this injustice. If it was reasonably likely that in 1963 a husband would have been advised that as far as the law was concerned he was entitled to sexual intercourse with his wife because there was no judicial decision to the contrary, the new law was not ascertainable. A court that changed the law, or at least made it ascertainable, should only do so prospectively, acquitting this defendant but making the law clear from then on. Enacted law is "law" if properly enacted, but if it is not reasonably ascertainable it will be unenforceable, whereas common law must be ascertainable in order to be "law". If enacted criminal law is not ascertainable a prosecution should be stayed: Christian v R (The Pitcairn Islands) [2006] UKPC 47 at [24] (not cited in PGA):

"Their Lordships would accept that the fact that a law had not been published and could not reasonably have been known to exist may be a ground for staying a prosecution for contravention of that law as an abuse of process... .",

The ECtHR has upheld convictions where retrospective change in the law has been reasonably foreseeable, as in the removal of the marital exemption defence to rape, upholding R v R [1992] 1 AC 599: SW and CR v UK (1996) 21 E.H.R.R. 363. It may be that with increasing focus on the idea of the rule of law this tolerance of injustice to the defendant whose prosecution marked the change in the law will diminish. The traditional view that judges declare the common law as it has been from time immemorial is now regarded as an outdated legal fiction.


Just as reasonable minds may differ on the application of known law to the facts of a case, so too may reasonable minds differ over how and when the common law may change. An obligation to obey law that has not yet been declared is a form of "common law" not within the definitions recognised by the majority in PGA. It is unjust.


There is plenty of authority for this, as Heydon and Bell JJ demonstrate in their dissenting judgments. A helpful summary from the House of Lords is in R v Rimmington [2005] UKHL 63 especially at [33] which concludes:


"There are two guiding principles: no one should be punished under a law unless it is sufficiently clear and certain to enable him to know what conduct is forbidden before he does it; and no one should be punished for any act which was not clearly and ascertainably punishable when the act was done. If the ambit of a common law offence is to be enlarged, it "must be done step by step on a case by case basis and not with one large leap": R v Clark (Mark) [2003] EWCA Crim 991, [2003] 2 Cr App R 363, para 13."

 There is a risk of cherry-picking authorities in this sort of argument, but I recommend Rogers v Tennessee, 532 U.S. 451 (2001) as an illustration of the upholding of a common law change but only where there had been a pervasive change in the law that should have put the defendant on notice that the law was clearly going to change. The change was not unfair or arbitrary. Scalia J dissented on the grounds that restrictions on legislative retrospective change should also apply to retrospective common law change. But even on the majority view there is a requirement of ascertainability before a defendant is bound by future decisions changing the common law.

Anyway, this interesting case raises questions about who defines the common law, is it just social mores or does it have to be declared by judges, whether retrospective effect should be more acceptable for common law offences than for legislated offences, and the extent to which social and legislative changes in some areas can influence assumptions about the common law in other areas. In New Zealand we have only one common law offence - some forms of contempt of court - but we do have common law defences.  It was not until as recently as 1986 that legislation removed the spousal immunity defence to rape, which implies that it did exist at common law (at least in a form that had been modified by earlier legislation) until then. Some people think we are behind Australia in our development, but - as one who can remember 1963 and the grassy knoll quite clearly - I don't think that in 1986 we were 23 years behind.

What is a verdict?

And for more on procedure as a means of giving effect to the substance of the law, see the failure of this concept in Blueford v Arkansas USSC No 10-1320, 24 May 2012.

A fundamental precept of the criminal law is that a jury will obey instructions from the judge. Here the jury were told the sequence required by the law of Arkansas for the deciding of verdicts on multiple and included charges. A decision on the most serious charge was required first, then the jury would proceed to the next serious, and so on until either they had reached a verdict of guilty, in which case they would stop there, or they had reached not guilty verdicts on each charge.

The jury had difficult reaching the end of this process,and the judge asked if they had decided any of the charges. The foreman said they had agreed that the defendant was not guilty on the two most serious charges. The judge told them to resume their deliberations and try to reach the end of the process. They were unsuccessful and they were discharged. On retrial the defendant was charged with all counts and pleaded previous acquittal on the two most serious ones.

The Supreme Court held by a majority that the special plea was not available. The jury had returned to the jury room and could have changed their minds on the most serious charges - they had not been asked whether they had. There were no verdicts returned at the trial.

Sotomayor J for the minority emphasised the importance of the rule against double jeopardy. She analysed the Arkansas process for deciding multiple charges of this kind and found it most unrealistic for the majority to say the jury might have changed their minds. In reality they had returned verdicts of not guilty on the two most serious charges.

True enough, the ritual incantations had not been followed in receiving the verdicts, but did that matter? What was the point of instructing the jury how to proceed, if that had no consequence because the technical question about verdicts was not asked by the judge? Had the jury decided the most serious charges? Should the prosecution have another opportunity to prove them? Did procedure serve the interests of justice here?

Too late! Or is it ...?

As everyone knows, procedural laws are for bringing the law into effect. Procedure brings the law to life, as I once said. Another thing that everyone knows is that senior appellate courts spend much of their time getting people to be sensible.

Lucaszewski v The District Court in Torun, Poland [2012] UKSC 20 (23 May 2012) gets people to be sensible about time limits for filing notices when statutory requirements are expressed in absolute terms.

The Court did not overrule Mucelli v Govt of Albania [2009] UKHL 2 (21 January 2009), noted here on 23 January 2009, although Lady Hale would have preferred to. However she agreed with the reasoning and conclusion of the leading judgment, delivered by Lord Mance.

Just as Mucelli reminded us of the Latin dies non, so does Lord Mance now remind us of  non constat [19]. These things don't go unnoticed.

Anyway, Lucaszewski allows that timely notice of intention to appeal an extradition order can be given although the document does not comply with the enacted form, if the court decides to permit the informality. Also, where a defendant is a citizen, as opposed to an alien, his civil right to enter and remain in the country attracts the right to a fair hearing when extradition is sought, and the court may extend time for the defendant to file documents if that would promote his rights, even if no discretion to extend time appears on the face of an enactment.

"[37] ...  I am not persuaded that the interests of finality and certainty outweigh the interests of ensuring proper access to justice by appeal in the limited number of extradition cases where this would otherwise be denied. There would not be "a reasonable relationship of proportionality between the means employed and the aim sought to be achieved".

Although this introduces a discrimination between aliens and citizens, which concerned Lady Hale, the more general point is that procedure must promote the aims of justice.

Causing moral responsibility

In Maybin v R, 2012 SCC 24 (18 May 2012) Karakatsanis J for the Court discusses the criteria for determining that D's acts have caused the actus reus, here the death of V, to an extent that criminal responsibility could be attributed to D. Essentially this is a moral question.

The legal requirement is that D's acts were a significant contributing cause of the death. There may be several causes of the death, but if D's contribution was more than de minimis, causation will be established for the purpose of deciding whether D is criminally responsible.

Legal responsibility  requires that D's acts had sufficient connection to that harm to warrant attribution of criminal responsibility [16]. There might be an insufficient connection if there was an intervening act so that D's acts were not a significant contributing cause of the harm. Such intervening acts might not sever the causal link between D's acts and the harm if they should have been foreseen by D at the time he acted [26], and this does not mean that D should have foreseen the exact details of the intervening act, but some sort of appreciation of what might happen, akin perhaps to the mens rea requirements of the offence, may be enough to make D responsible [36], [38].

The independent acts might not sever the causal connection between D's acts and the harm if, although those independent acts were unforeseeable, the relative weights of the causes of the harm [46] were such as to leave a more than minimal causal link attributable to D.

In any event, these are just analytical aids, not legal tests, for determining causal responsibility. In different factual matrices they might point in different ways. The fundamental question is always whether D's acts were a significant contributing cause of the harm.

The moral question whether D should be held responsible when there are multiple causes of a harm will often attract different answers, as is illustrated in the differences of opinion in the Court of Appeal here. There had been a fight in a bar and D had knocked out V, who was then struck by a third person, a bouncer. V died and D was charged with manslaughter. You be the judge (after reading the full facts in the case, of course).

Suppressing contempt

Our Court of Appeal has today confirmed the inherent power of a court to suppress its judgments on an interim basis in order to protect the defendant's right to a fair trial: Siemer v Solicitor-General [2012] NZCA 188.

This is a departure from Privy Council and House of Lords decisions: Independent Publishing Ltd v Attorney-General of Trinidad and Tobago [2004] UKPC 26, [2005] 1 AC 190, Attorney-General v Leveller Magazine Ltd [1979] AC 440. Since the establishment of our Supreme Court, decisions of the Privy Council from other jurisdictions are of persuasive authority only [37].

The Court of Appeal held [91] that the same position on suppression orders applies under the Criminal Procedure Act 2011 ("CPA"). I will consider the other aspects of the judgment by reference to what the position should be after commencement of that Act.

Of equal interest is the Court's discussion of contempt proceedings. Sometimes contempt is an offence. This is when an enactment creates the offence and provides for a punishment. Provisions doing that, however, provide that they do not limit the power of a court to punish for contempt: see s 365(3) of the CPA.

Contempt that is punishable pursuant to the court's inherent power is probably not an "offence" although there is no definition of "offence" in the CPA except in the schedule where a definition is inserted in another act. In any event, the Court of Appeal accepted [96] that contempt proceedings outside those defined in an enactment are neither criminal nor civil proceedings, and they are not prosecuted by the laying of an information (nor, by implication, by the filing of a charging document under the CPA). Although the criminal standard of proof applies, these sorts of contempt do not result in a conviction but rather in a "finding" of contempt. Such a finding, and the sentence, may be appealed under the CPA, s 260.

If contempt outside the statutory definition is an offence, proceedings in respect of it should be commenced under the CPA by the filing of a charging document. However it is more likely that it is not an offence and the Solicitor-General commences proceedings for this sort of contempt by the filing of an originating application and supporting affidavits, and the defendant may answer by affidavit [93]. The criminal procedure of entering a plea does not apply.

In Siemer the appellant had published a judgment that the High Court had temporarily suppressed. That suppression order was designed to protect the right of defendants in a notorious criminal case to fair trials. As a result of losing the appeal the defendant was ordered to commence serving his sentence of six weeks in prison.

Dicta in this case [68]-[72] on the right to a fair trial are significant too. The defendants in the criminal case had the right to a fair trial and the court used its inherent power to regulate its procedures for the ultimate objective of securing fairness. However Mr Siemer, who was not subject to criminal proceedings but only to the "neither criminal nor civil" proceedings for contempt, did not have the protection of the rights given by the New Zealand Bill of Rights Act 1990 to people who are charged with offences [95]. But he did have the right to justice, enjoyed by everyone pursuant to s 27 of the Bill of Rights, which applies the rules of natural justice to the determination of his rights, obligations or interests.

My 500th posting: Probative value and independent evidence

Probative value is the tendency that an item of evidence has to prove an issue in a case. Does the measure of that tendency depend on other independent evidence in the case?

For example, consider these facts on a charge of murder:

"[2] ... A hair found on the deceased's thumbnail had been subjected to mitochondrial DNA testing. The results of that testing showed two things: first, that the appellant could have been the donor of the hair and, second, how common the DNA profile found in the hair was in the community. This second aspect of the results was expressed in evidence both as a frequency ratio [sometimes called a "random occurrence ratio" or a "frequency estimate"] and as an exclusion percentage. The expert who had conducted the test gave evidence to the effect that one in 1,600 people in the general population (which is to say the whole world) would be expected to share the DNA profile that was found in the hair (a frequency ratio) and that 99.9 per cent of people would not be expected to have a DNA profile matching that of the hair (an exclusion percentage)."

The case, Aytugrul v R [2012] HCA 15 (18 April 2012) concentrated on whether expressing the statistics as an exclusion percentage was improperly prejudicial. It was argued for the appellant that to say 99.9% of people wouldn't match suggests that it is 99.9% likely that the appellant was guilty.

In the circumstances of this appeal it was unanimously held that there was no improper prejudice because the jury had been carefully directed on how they should reason. Determining whether there is improper prejudice requires having regard to the whole of the evidence [30].

The question I started with did not need to be answered in this case because the appeal turned on whether there was illegitimate prejudice, but Heydon J considered it in some detail [41]-[65], especially in relation to whether giving evidence of one way of expressing the scientific results rendered an alternative and equivalent way redundant and inadmissible.

Evidence about the matching of samples is like all other evidence: it raises the question, given the existence of this evidence, what is the probability that the defendant is guilty? This probability of guilt is the ultimate issue and involves consideration of all the admissible evidence in the case that the fact-finder accepts as true. The probative value of an item of evidence is its tendency to prove an issue, and where that is the ultimate issue, the question arises as to whether probative value should be assessed by considering the item of evidence in isolation, or whether its probative value should be considered in the context of other evidence.

Logicians would be isolationists but with some qualification. Independent evidence is irrelevant but dependent evidence is relevant to the probative value of an item of evidence, as is contextual evidence.

Usually evidence of the sort of match in Aytugrul would be given in terms such as: "whatever the probability that the defendant is guilty based on the other evidence in the case, this evidence increases that probability by 1600 times." Sometimes the same thing might be said in words rather than numbers, for example by saying the evidence strongly supports the proposition that the defendant is guilty (see [32]).

I am assuming that the ratio would result in a figure of 1600, although that is not clear. There might (here I am not referring to the particular facts of this case) be innocent explanations for the evidence of a match, even if the defendant was guilty: the test result might have been a false positive so that in reality it was someone else's hair, or the defendant's hair might have been transferred there when the police had custody of the body, or there might have been an error in handling scene exhibits and a mix-up between what was found on the victim and medical samples from the defendant. The probability of those sorts of errors, although normally small (but if larger they could make the test result irrelevant - see my concluding comment below), should be taken into account. In that sense some other evidence in the case could be relevant to the probative value of the evidence of a match. But that other evidence would not be independent of the existence of the hair.

Independent evidence, such as evidence of the time at which the defendant and the victim were together, and the time of death, does not affect the probative value of the evidence of the match between the hair and the defendant. It does not affect the numerator or the denominator of the likelihood ratio. Those are based on assumptions, or "givens": that the prosecution hypothesis of guilt is true, and that the defence hypothesis of innocence is true.

I infer from the extract from Dr Buckleton's evidence at [16], where he mentioned the lesser of two evils, that he would prefer not to put his evidence as either a random occurrence ratio or an exclusion percentage. The reference in the judge's summing up to Dr Buckleton's evidence of expecting the occurrence of that sort of hair in the relevant population of innocent people as between one in 50 people and one in 100 or less, seems to be a reference to the denominator of the likelihood ratio being between 0.01 and 0.02. Dr Buckleton usually gives his results as a likelihood ratio following the Bayesian approach, and he supports the "Bayes and the Law" site. Plainly there was a discrepancy between the experts in Aytugrul as to the statistical values. In any event, in this case Dr Buckleton said that the test results did not indicate a match: Aytugrul v R [2010] NSWCCA 272 at [55]. The implication of that evidence seems to have been lost in this case.

Significantly, the High Court recognised that the question of how this sort of evidence should be expressed is a question of psychology upon which the Court would need further evidence before making a legal rule about that [22]-[24]. (I recommend to anyone interested in this sort of thing the Nobel laureate psychologist Daniel Kahneman's "Thinking, Fast and Slow".) The Court did not say that evidence of this kind must be given in the form of a likelihood ratio, and it specifically did not lay down a rule that it could never be given as an exclusion percentage [21].

Note the difference between the probative value of evidence and the effect of its combination with other evidence in the case on the ultimate probability of guilt. In BBH v R, discussed here on 31 March 2012, it was this latter combination that was relevant in considering the Pfennig requirement.

There is a need for experts to agree on how to express their conclusions when giving evidence, and for legal professionals and judges to become familiar with correct reasoning with probabilities. In the NSWCCA in this case (see above link) McLellan CJ at CL, whose judgment is well worth reading for its references to specialist articles - including a paper by Kahneman - on the effects on juries of different ways of expressing statistical results, held that the trial was unfair, notwithstanding that the verdict was, on the evidence, not unreasonable, because the jury may have been led to think that the exclusion percentage was the same as the percentage likelihood of the defendant being guilty. The other judges in that court regarded the statistical evidence as strongly supportive of guilt and that was sufficient, there being no identifiable unfair prejudice to the defendant (Simpson J at [193]-[195]).

Indeed, as Bayesians know, if the other evidence in the case only established guilt to a probability of 0.1, the hair match evidence would increase that to 0.99. The same is true without any other evidence, but obviously there have to be other proven facts to give the match evidence context in relation to the offence charged. It is also appropriate to talk about groups of facts which together have probative value in relation to an issue, but their individual probative value is different from their collective probative value. Again, context is relevant, and it may be inappropriate to consider the probative value of a fact in isolation.

So what would have been unfair prejudice here? If the jury had not found as a fact that the hair had the DNA profile that the prosecution claimed it had, the other evidence about the frequency of a match would be irrelevant. To find as a fact that the hair had the alleged DNA profile, the jury would have to be satisfied about that to the standard of the balance of probabilities. Most subsidiary facts - that is, facts that are not elements of the relevant offence - only need to be proved to that standard, although there is an argument, popular - to the point of being law - in Australia, that critical facts have to be proved beyond reasonable doubt. Be that as it may, unfair prejudice would have existed if the match frequency evidence could have distracted the jury from its proper assessment of whether the hair did indeed have the DNA profile claimed for it.